Engineering Slowly Digestible Starch via Physical and Enzymatic Modification: Structural Mechanisms of Formation

ABSTRACT

Dietary starch varies in digestion rate: Rapidly digestible fractions cause sharp postprandial glucose spikes linked to chronic disease, whereas slowly digestible starch (SDS) breaks down gradually, producing a moderate glycemic response. Increasing SDS content is therefore a key objective in developing carbohydrate‐rich foods with improved metabolic functionality. Physical, enzymatic, and combined modification strategies have been reported to enhance SDS, but outcomes are often inconsistent, with the same treatment increasing SDS under some conditions while favoring rapidly digestible or resistant fractions under others. The structural basis of this variability, and how conditions can be adjusted to reliably favor SDS formation, remains insufficiently synthesized. This review compiles studies reporting quantified SDS increases relative to unmodified controls following physical, enzymatic, and combined strategies, linking these changes to structural and functional transformations, and addressing conditions under which SDS declines. SDS formation is positioned as crystallization arrested at nucleation, distinct from the chain maturation yielding resistant starch, with chain length, branching density, and granule or matrix architecture governing the balance among SDS, resistant, and rapidly digestible fractions. These insights, together with strategies for recovering SDS when treatments underperform, aim to inform rational design of starch‐based foods with tailored digestibility for glycemic management and to help close the gap between laboratory‐scale research and commercial translation.

1. Introduction

Starch constitutes the predominant carbohydrate component of human diets and is a major determinant of postprandial glycemic responses. The rate and extent of starch digestion influence glucose absorption kinetics, insulin secretion, and long‐term metabolic health. On the basis of in vitro digestion behavior, starch is typically categorized into rapidly digestible starch (RDS), slowly digestible starch (SDS), and resistant starch (RS) (Englyst et al. 1992; Freitas et al. 2025). RDS is hydrolyzed quickly in the oral cavity and upper small intestine, causing a rapid rise in blood glucose within the first 20 min of digestion. Unlike RDS, SDS is hydrolyzed more gradually (20–120 min), with pancreatic α‐amylase acting in the duodenum and mucosal brush border enzymes completing hydrolysis along the jejunum and ileum, leading to a sustained release of glucose, improved glycemic stability, and reduced insulin demand (Magallanes‐Cruz et al. 2017; Miao et al. 2015; Rostamabadi et al. 2024). In contrast, RS escapes digestion in the small intestine and undergoes fermentation in the colon. The physiological relevance of SDS lies in its ability to attenuate postprandial glucose spikes, supporting better metabolic control (Doan et al. 2025; Miao et al. 2015). Increasing evidence suggests that diets enriched in SDS may contribute to reduced glycemic variability, improved metabolic regulation, and enhanced satiety (Freitas et al. 2025; Miao et al. 2015). Beyond its nutritional role, SDS also offers potential in controlled delivery systems, where its gradual digestion can be used to modulate the release of bioactive compounds within complex food matrices. These physiological benefits have stimulated extensive research aimed at developing technological approaches capable of increasing the SDS proportion in starch‐based foods (Singh et al. 2010).

Previous studies have shown that the formation of SDS is not solely governed by intrinsic starch properties but is also strongly influenced by extrinsic factors, particularly processing conditions (Magallanes‐Cruz et al. 2017; Miao et al. 2015). Intrinsic factors include molecular and granular characteristics such as amylose–amylopectin ratio, chain length distribution, degree of branching, crystalline type, and granule architecture. These features determine the extent of molecular packing, double‐helix stability, and enzyme penetration, thereby regulating the inherent susceptibility of starch to enzymatic hydrolysis (Raghunathan et al. 2025). However, these intrinsic characteristics are often modified or restructured during processing, which can significantly alter starch digestibility behavior.

Processing conditions, such as temperature, moisture content, shear forces, and enzymatic activity, can disrupt native granular organization, induce gelatinization, and promote subsequent molecular rearrangements during cooling or storage. These transformations include reassociation of starch chains, formation of double helices, development of imperfect or perfect crystallites, and changes in the connectivity between amorphous and crystalline regions. As a result, processing can either enhance or reduce enzyme accessibility depending on the extent and nature of structural reorganization (Chi et al. 2022). In particular, controlled processing can promote the formation of semi‐ordered structures and compact matrices that restrict the diffusion of digestive enzymes and limit substrate availability. For example, treatments such as heat–moisture conditioning, annealing, extrusion, and enzymatic debranching can increase molecular packing density, stabilize double‐helical structures, and generate slowly hydrolysable domains within the starch matrix (Huang et al. 2015; Xu et al. 2018). In addition to molecular‐scale changes, processing also influences supramolecular and granular features, including pore structure, surface morphology, and particle size, which further regulate enzyme penetration and hydrolysis kinetics.

SDS naturally occurs in a range of food sources; however, its content varies considerably depending on botanical origin and structural characteristics. Foods, such as minimally processed cereal products (e.g., pasta, durum wheat products), legumes (e.g., lentils, chickpeas), and some whole grains, have been reported to contain moderate levels of SDS, typically ranging from approximately 10%–40% of total starch, depending on variety and processing conditions (Chisbert et al. 2024; Dodi et al. 2023; Farooq and Yu 2024). In particular, products with compact structures or higher amylose content tend to exhibit greater SDS fractions due to restricted enzyme accessibility. Similarly, certain traditional foods prepared under controlled conditions (e.g., al dente pasta or parboiled rice) may retain higher SDS levels because of limited gelatinization and enhanced structural integrity (Dodi et al. 2023; Liu et al. 2021; Tian et al. 2018).

Beyond food matrix and processing format, intrinsic genotypic variation within a given crop species further contributes to the natural range of SDS content. Because cultivar‐driven SDS differences are typically evaluated at the whole food level (e.g., whole flour) in the literature, the data reflect starch within its native food matrix rather than isolated starch. This native SDS pool differs markedly in how well it survives cooking. In rice, raw‐flour SDS across genotypes spans 25.3%–36.5%, but cooking sharply erodes this fraction, even the highest retaining genotypes fall to only 7.69%–8.14% SDS after cooking, versus 1.2%–2.8% in other accessions, indicating that genotype affects how much SDS survives heat treatment (Lee et al. 2025). A similar genotype‐driven range is reported in wheat, where SDS in raw whole‐meal flour ranges from 11.87% to 27.91% across 272 varieties and correlates with kernel hardness and total starch content (Du et al. 2025), though whether this native variability survives processing to the same extent as in rice remains to be confirmed. Pseudocereals illustrate an even sharper contrast: raw amaranth starch can reach approximately 51% SDS, among the highest of any grain, yet its small granule size and low amylose content cause this structure to collapse almost completely upon heat treatment, leaving amaranth‐based cooked products with minimal SDS or RS (<1%) (Srichuwong et al. 2017). Legumes show a similar vulnerability: Raw lentil and chickpea starch is dominated by the SDS fraction, but conventional boiling reduces SDS by 64%–74% relative to raw flour (Dhull et al. 2023), whereas nonthermal high‐pressure processing increased the SDS content of chickpea from 50.5% to 60.9% while reducing RDS (Karunarathna et al. 2024). These contrasting responses indicate that botanical source and genotype determine the inherent potential for SDS formation, whereas the processing method and conditions govern the extent to which this potential is preserved, enhanced, or lost in the final food product.

Common processing operations, such as excessive heating, high moisture treatment, milling, and prolonged cooking, can disrupt starch structure, promote gelatinization, and increase enzyme accessibility, thereby shifting the balance toward RDS and reducing SDS content (Alvarado et al. 2026). As a result, many widely consumed starch‐rich foods exhibit relatively low SDS levels despite their starch content (Teng et al. 2025). This highlights the need for targeted processing strategies that can preserve or enhance SDS formation. Therefore, identifying and optimizing physical and enzymatic modification methods to increase SDS in food systems represents a critical area of research for the development of starch‐based products with improved metabolic functionality.

Although starch digestibility has been widely investigated, much of the existing research has primarily examined overall changes in digestibility fractions (i.e., RDS, SDS, and RS) in response to processing, rather than specifically focusing on the enhancement of the SDS fraction. Consequently, studies reporting increases in SDS are distributed across diverse processing approaches and experimental systems, and SDS is often reported only as part of broader digestibility outcomes, making it difficult to directly compare treatment conditions, magnitude of change, or identify common structural mechanisms responsible for its formation (Wongsagonsup et al. 2026). Individual studies have linked increases in SDS to specific structural modifications, such as changes in crystallinity, chain‐length distribution, or matrix organization, but these relationships are rarely synthesized across different modification strategies, leaving a consolidated understanding that connects processing conditions, structural reorganization, and measurable SDS increases largely lacking. Given the growing interest in SDS for developing functionally optimized foods, there is a need to organize and integrate the available evidence in a structured and comparable manner (Wongsagonsup et al. 2026; Xiao et al. 2021). Therefore, this review focuses on compiling and synthesizing studies that have demonstrated quantified increases in SDS relative to unmodified controls, with reported pre‐ and posttreatment values and, where available, statistical validation following physical, enzymatic, and combined modification strategies. Emphasis is placed on reporting treatment conditions, quantifying changes in SDS, and linking these changes to underlying structural and functional transformations. By organizing the available evidence across different modification approaches, this review aims to provide a clear framework that supports the understanding and practical application of strategies for enhancing SDS in starch‐based systems.

In this review, SDS is defined based on the widely adopted in vitro digestibility classification, whereby starch fractions hydrolyzed between 20 and 120 min are categorized as SDS (Englyst et al. 1992). Because methodologies are not fully standardized across the literature, direct comparison between studies should be made with caution. Priority is therefore given to studies that explicitly report SDS values before and after treatment. Where studies report other related digestibility indicators such as RDS, RS, hydrolysis index, or estimated glycemic index, these are discussed only as supporting evidence for structural or mechanistic interpretation and not as direct substitutes for SDS.

2. Physical Modification Strategies and Their Structural Implications

Physical modification strategies, including thermal–moisture treatment, cold plasma, ultrasound, dry heating, high‐pressure treatment (HPT), heating and cooling, freezing, and microwave processing (Figure 1 and Table 1), represent widely investigated approaches for modulating starch digestibility, as they alter starch structure without introducing chemical substituents. These treatments primarily affect the organization of crystalline and amorphous regions within starch granules, thereby influencing hydrolysis behavior (Gou et al. 2019; Wang et al. 2022).

FIGURE 1.

Schematic illustration of the effects of physical modification methods on starch structure and digestibility. HMT, heat–moisture treatment; SDS, slowly digestible starch.

TABLE 1.

Representative examples of single physical modification treatments and their influence on slowly digestible starch formation.

Starch source	Modification	SDS before (%) a	SDS after (%) a	Absolute change in SDS (%) b	Structural observations	Reference
Wheat	High‐pressure treatment (300 and 600 MPa; 30 min, room temp. starch–water suspension of 20%)	13.50	18.20	+4.70	Reduced swelling power and viscosity	Bajaj et al. (2022)
Potato starch	High‐pressure treatment (300 and 600 MPa, 30 min, room temp., starch–water suspension of 20%)	31.20	35.20	+4.00	Reduced relative crystallinity	Bajaj et al. (2022)
Sorghum	High‐pressure treatment (600 MPa, 30 min, room temp., 20% moisture content)	36.90	45.10	+8.20	Granules deformed and fused; fissures and deep holes on the surface; decreased relative crystallinity; A‐type transformed to B‐type	Liu et al. (2016)
Potato	Planetary ball milling (400 rpm; 5 h on dry starch powder)	1.95	50.00	+48.05	Partial disruption of semicrystalline regions converted RS into SDS while increasing enzyme accessibility	Lv et al. (2019)
Yam	Annealing (at 1:2 and 1:4 starch to water ratio and 50°C for 24 h)	13.35	24.18	+10.83	Increased gelatinization temperature and molecular ordering	Bora et al. (2023)
Proso millet	Annealing (12 g starch in 60 mL distilled water was heated in a water bath at 50°C for 24 h)	5.09	12.03	+6.94	Decreased relative crystallinity and short‐range molecular order	Wang, Ye, et al. (2023)
Potato	Heat–moisture treatment (30%–35% moisture, 100–120°C, 2 h)	12.65	34.41	+21.76	Reduced relative crystallinity and induced internal molecular rearrangement	Brahma and Sit (2020)
Proso millet	Heat–moisture treatment (15% moisture, 100°C, 2 h)	5.09	11.69	+6.60	Reduced crystallinity and short‐range order with molecular rearrangement and partial hydrogen‐bond reorganization, increasing SDS	Wang, Ye, et al. (2023)
Foxtail millet	Heating–cooling (gelatinized starch, cooled at 24°C for 30 min and 5 days)	14.20	18.33	+4.13	Gelatinization increased starch accessibility, whereas subsequent retrogradation promoted amylose and amylopectin recrystallization, shifting starch digestibility toward RS rather than SDS	Xing et al. (2024)
Maize starch	Heating–cooling (cooked as a 10% starch paste to fully gelatinize, then stored at 4°C for 40 h)	∼13%–15%	27–28	+13–14	Enhanced short‐range helical order and aggregate density	Zhang et al. (2008)
Maize	Microwave treatment (30% moisture content; 6.63 W/g for 5 min)	28.64	47.16	+18.52	Greater molecular reorganization and increased formation of ordered structures markedly increased SDS, although RS decreased because of enhanced starch gelatinization	Xu et al. (2019)
Potato	Microwave treatment (30% moisture content; 6.63 W/g for 5 min)	23.12	42.03	+18.91	High‐power microwave induced substantial structural modification and molecular rearrangement, increasing SDS while reducing RS due to greater starch gelatinization	Xu et al. (2019)
Lotus root	Cold plasma (40% moisture content; 0.8 A current; 40 V, voltage; 12 mm discharge distance; and air as a gas; treatment time: 90 s)	8.17	12.67	+4.50	Surface oxidation, increased molecular ordering, reduced enzyme accessibility	Sun, Sun, et al. (2023)
Corn	Cold plasma (15% moisture content; compressed air at a flow rate of 15 cm3/min, and the power of 120 W, time: 21 min)	4.18	6.66	+2.48	Cross‐linking and amylopectin chain rearrangement produced more ordered, enzyme‐resistant structures, increasing SDS	Liu, Zhang, et al. (2025)
Potato	Cold plasma (15% moisture content; compressed air at a flow rate of 15 cm3/min, and the power of 120 W, time: 21 min)	5.68	11.59	+5.91	Greater plasma penetration into the loosely packed B‐type granules promoted amylopectin depolymerization and cross‐linking, increasing the formation of slowly digestible structures	Liu, Zhang, et al. (2025)
Pea	Ultrasound (30%, w/w slurry, 28 kHz, 300 W, 30 min; 5 s on and 5 s off, 25°C)	41.00	44.00	+3.00	Ultrasonic disruption of the granular, crystalline, and double‐helical structures reduced RS and converted part of the resistant fraction into slowly digestible starch, increasing SDS	Xiao et al. (2021)
Waxy rice	Freeze‐thawing (10 cycles, −20°C, thawing for 2 h at 25°C)	23.80	50.30	+26.50	Anchor‐point chains (DP 9–30) associate early, restricting enzyme access to the attached long chains (DP > 30), forming a semi‐ordered structure that slows digestion without full resistance	Tao et al. (2015)
Ginger starch	Freeze‐thawing (10% starch suspension gelatinized at 100°C for 20 min, followed by freeze–thaw cycles (−20°C for 2 h → 50°C for 30 min); repeated 1–7 cycles)	11.73	20.27	+8.54	The initial freeze–thaw cycle promoted starch retrogradation and partial molecular rearrangement, forming ordered double‐helical/crystalline regions that slowed enzymatic digestion	Wang, Liang, et al. (2023)
Maize	Dry heat treatment (130°C for 3 h, moisture content dried below 10% beforehand)	53.05	62.48	+8.98	Dry heat at low moisture partially reorganizes starch chains without destroying the granule, forming a moderately ordered structure that resists digestion more than native starch but less than the fully resistant fraction	Zou et al. (2020)
Taro	Dry heat treatment (1, 3, 5, and 7 h at 130°C; 90°C, 110°C, and 130°C for 5 h)	16.43	31.30	+14.87	Reduced relative crystallinity; degraded starch; increased short‐range molecular order	Hui et al. (2024)
Rice	Twin‐screw extrusion; solid feed 10 kg/h, water feed 2.5 L/h, screw speed 90 rpm, barrel temperature profile 70°C → 80°C → 90°C → 100°C → 100°C (feed zone to die)	3.52	5.32	+1.80	Disrupted semicrystalline structure followed by partial chain reorganization into moderately ordered structures, increasing SDS	Ma et al. (2026)
High‐amylose maize starch	Twin‐screw extrusion, screw speed 250–600 rpm, barrel temperature 100–160°C, feed moisture 16.4%–22.5%	20.40	27.50	+7.10	Partial survival of native granular/crystalline structure, retrogradation on cooling, and amylose‐lipid complex formation (new DSC endotherm near 130°C) together preserved slow‐digesting character	Robin et al. (2016)
Germinated corn starch	Static magnetic field (30 mT for 2 h)	8.58	12.49	+3.91	Increased porosity; enhanced relative crystallinity and molecular order	Jiang et al. (2025)
Pre‐gelatinized cassava starch	Alternating magnetic field (7–8 mT, 250 Hz, 25°C, 10 min	36.99	41.03	+4.04	Enhanced crystallinity and short‐range molecular order through amylose rearrangement and recrystallization, reducing enzymatic hydrolysis	He, Du, et al. (2025)

Note: Comparisons across studies should be interpreted cautiously because digestion protocols, sample preparation, and starch fractionation methods may differ.

Abbreviations: DP, degree of polymerization; RS, resistant starch; SDS, slowly digestible starch.

^a SDS values are reported as presented in the cited studies.

^b Absolute change in SDS was calculated as SDS after–SDS before (%).

The effectiveness of these approaches in enhancing SDS varies considerably across starch systems, reflecting differences in botanical source and processing conditions (Su et al. 2020). The extent and nature of structural modification, and consequently SDS formation, depend on the type and severity of the applied physical treatment, as discussed in the following sections.

2.1. High‐Pressure Treatment

HPT, also referred to as high‐hydrostatic pressure processing, is a nonthermal physical modification method in which starch or starch‐containing systems are exposed to elevated pressures, commonly in the range of 100–600 MPa, for controlled holding times and at defined water contents or slurry concentrations (Wongsagonsup et al. 2026). As an attractive nonthermal technology, HPT enables the creation of unique textural attributes while preserving nutritional components and avoiding thermal degradation (Bajaj et al. 2022). Its effect on digestibility is highly condition‐dependent because pressure can either induce limited molecular rearrangement that favors SDS formation or drive extensive pressure‐induced gelatinization that increases starch susceptibility to enzymatic attack (Zheng et al. 2025). Recent investigations emphasize that the impact of the treatment relies heavily on the botanical source and internal interactions between starch chains, particularly the amylose‐to‐amylopectin ratio (Bajaj et al. 2022; Rong et al. 2024).

Available literature (examples in Table 1) indicates that moderate‐to‐high pressures, when applied under appropriate conditions, can increase SDS by promoting partial structural disruption followed by molecular reorganization. The SDS content in various starches, including wheat, corn, waxy corn, potato, sweet potato, kidney bean, and rice starch, increased with the increase in pressure from 300 to 600 MPa, whereas RDS fell and relative crystallinity declined in parallel across the same range, indicating a consistent pressure‐dependent response rather than an effect confined to particular starch types (Bajaj et al. 2022). The response to HPT varied considerably among botanical sources. Wheat starch showed the largest absolute increase in SDS, rising by 4.7 percentage points from 13.5% to 18.2%. In contrast, kidney bean and sweet potato starches, which initially contained very little SDS, showed the greatest proportional increases, from 2.8% to 5.9% and from 1.3% to 4.6%, respectively, although their final SDS contents remained relatively low. This pattern indicates that the magnitude of the HPT response is not predicted by initial SDS content alone, and that botanical source independently governs how starch responds to treatment (Bajaj et al. 2022).

The increase in SDS following HPT may result from pressure‐induced disruption of amylopectin double helices and increased penetration and redistribution of water and other components within starch granules. These changes partially disrupt the highly ordered, enzyme‐resistant structures associated with RS, converting some of this fraction into less resistant structures that are digested more slowly than RDS and are, therefore, classified as SDS. However, more extensive structural disruption may further increase enzyme accessibility, shifting starch toward the RDS fraction (Bajaj et al. 2022; Okur et al. 2019). Nuclear magnetic resonance analysis suggests that high pressure drives water molecules into cornstarch granules, where they become entrapped within crystalline regions, increasing hydration and facilitating structural reorganization into slowly digestible domains (Okur et al. 2019). Consistent with this mechanism, increasing pressure followed by controlled retrogradation increased the SDS content of waxy wheat starch from approximately 18% to 30%. However, RDS also increased while RS decreased, indicating that HPT simultaneously disrupted highly enzyme‐resistant structures and promoted the formation of less ordered structures with intermediate resistance to enzymatic digestion (Hu et al. 2017).

These SDS gains were accompanied by measurable functional trade‐offs that held broadly across the starches tested: swelling power declined with increasing pressure in various pulse and cereal starches, relative crystallinity declined in all seven, and peak and final pasting viscosities declined in most sources as well, with waxy corn starch as the consistent exception and potato starch also retaining its final viscosity under pressure. This indicates that pressure‐induced SDS formation is not functionally neutral but coincides with altered processing‐relevant properties across botanical sources, even though the degree of trade‐off, like the degree of SDS gain itself, varies by starch type (Bajaj et al. 2022).

The findings indicate that the RDS–SDS–RS response to HPT depends on an interaction between botanical source (and its underlying granule crystallinity and structure) and specific processing parameters, including pressure magnitude, holding time, and whether a subsequent retrogradation step is applied, rather than on pressure magnitude alone.

A key limitation across many studies is that starch digestibility is assessed only immediately after treatment. Although some studies have monitored changes in gel firmness and crystallinity during refrigerated storage, RDS, SDS, and RS are rarely remeasured (Bajaj et al. 2022). Therefore, the stability of HPT‐induced increases in SDS during subsequent storage remains unclear.

2.2. Thermal–Moisture Treatments

Thermal–moisture treatments, including heat–moisture treatment (HMT) and annealing (ANN), are widely applied physical modification strategies that alter starch structure through controlled heat and moisture without causing complete gelatinization. Under these conditions, limited molecular mobility is induced within the granule, facilitating the rearrangement of amylopectin chains and the development of more stable crystalline domains while largely preserving granule integrity (Ye et al. 2026). Water acts as a plasticizer during these processes, and this restricted molecular movement allows chains to realign into denser, more heat‐resistant architectures (Chung et al. 2009; Zeng, Chen, et al. 2015). These structural changes are commonly linked to reduced enzymatic accessibility and, often, higher SDS content. This is not the only route to SDS gain reported, however: At least four distinct pathways, reduced enzyme accessibility, branch‐point loosening that increases accessibility, short‐chain recombination, and thermally stable amylose‐lipid or starch‐protein complex formation, have each been independently linked to SDS formation under thermal–moisture treatment, depending on starch source, amylose content, and treatment protocol.

2.2.1. Heat–Moisture Treatment

HMT enhances SDS under optimized conditions across several starch sources. Sweet potato starch treated at 100°C and 25% moisture increased its SDS fraction from 28.27% to 39.59% (Sun, Qin, et al. 2023), and japonica rice starch reached a peak SDS of 38.09% after 6 h of treatment, up from 34.38% (Zhang et al. 2024). In low glycemic index rice starch, autoclave‐based HMT at 100–120°C increased SDS from 24.5% to 26.8% while more than doubling RS, from 6.9% to 15.8%, an outcome the authors attributed to increased particle size and surface compaction restricting enzyme access, formation of thermally stable V‐type amylose‐lipid complexes, and a starch‐protein composite matrix that partially persisted even after full gelatinization (Zhou et al. 2026).

Previous studies on HMT have also highlighted the importance of amylose content in starch selection, reporting that lower amylose starches may be more favorable for SDS formation. Under identical HMT conditions at 100°C, the SDS content of normal potato starch containing 40.68% amylose increased from approximately 20% to 25% (+5 percentage points), whereas that of waxy potato starch containing only 3.40% amylose increased from approximately 25% to 37% (+12 percentage points). This difference was attributed to amylose and long amylopectin chains preferentially forming stable double‐helical structures associated with RS in normal starch, whereas interactions among shorter amylopectin chains in waxy starch favored the formation of structures within the SDS fraction (Fang et al. 2023).

Increasing HMT severity does not necessarily lead to a continuous increase in SDS. In both normal and waxy potato starches, SDS increased to an optimum level, at 100°C for normal starch and 110°C for waxy starch, but subsequently decreased when the treatment temperature reached 120°C. This decline was accompanied by an increase in RS, suggesting that more severe HMT promoted further molecular reorganization and converted part of the SDS fraction into more stable, enzyme‐resistant structures classified as RS (Fang et al. 2023). Wheat starch showed a comparable pattern when treated at 120°C for 2 to 12 h, with SDS peaking at 48.36% after five cycles before declining slightly to 48.07% at six, and continuous treatment peaking at 46.94% at 10 h before declining to 46.21% at 12 h (Su et al. 2020). Therefore, SDS formation under HMT follows an optimum in treatment severity and duration, beyond which further reorganization favors RS over SDS, and this optimum varies with starch source.

In terms of treatment duration, extending HMT does not necessarily produce the same effect as applying repeated treatment cycles, because intermittent heating may promote more effective molecular reorganization. For example, a single HMT cycle applied to wheat starch at 120°C and 30% moisture for 2 h increased SDS from 21.87% to 43.15% (+21.28 percentage points), whereas RS decreased from 28.50% to 15.48%. Repeating the treatment for up to five cycles further increased SDS to a maximum of 48.36% (+26.49 percentage points). Notably, repeated‐cycle HMT consistently produced more SDS than continuous HMT of an equivalent total duration, indicating that treatment pattern, rather than duration alone, is important for maximizing SDS formation (Su et al. 2020).

Across previous studies, HMT consistently altered pasting behavior and, in most sources, reduced relative crystallinity. Swelling power and solubility responses diverged between starches, a difference attributed to amylose content. Digestibility outcomes moved broadly in the same direction, with RDS decreasing and SDS and RS increasing, but the relative contribution of each fraction varied by source, and in some cases both SDS and RS rose and then declined with extended treatment time. Overall, HMT reliably lowers digestibility, but the structural pathway and the fraction primarily responsible, SDS or RS, depend on starch source and treatment conditions (Su et al. 2020; Zhou et al. 2026). Most HMT studies measured digestibility only immediately after treatment, and information on the stability of SDS after further storage or subsequent treatments remains scarce. HMT requires substantially less energy and no specialized equipment and is broadly low cost and effluent‐free relative to chemical modification, though typical processing durations of 1–24 h remain a scalability consideration (Fang et al. 2023; Fassinou et al. 2025).

2.2.2. Annealing (ANN)

ANN operates through the same moisture‐mediated crystalline reorganization described above, typically at lower temperatures and considerably longer durations than HMT, and has produced SDS gains across several pure starch sources (Chung et al. 2009). In common wheat starch, eight repeated ANN cycles increased SDS from 44.99% to 48.14% (Su et al. 2020). Rice starch annealed at 50°C for 48 h showed SDS rise from 24.16% to 29.32% (Xiang et al. 2022). One step ANN of Castanopsis sclerophylla starch for 24 h lifted SDS from 18.3% to 23.0%, with RS unchanged despite increased crystallinity, an improvement attributed to stabilizing starch crystallites and strengthening amylose–amylopectin interactions (Shin et al. 2005). In sweet potato starch, ANN at 55°C and 50% moisture for 12 h increased SDS from 15.6% to 31.0%, an absolute gain of 15.4 percentage points and an approximately 2‐fold increase (Shin et al. 2005).

Not every study finds this direction. In corn, pea, and lentil starch, ANN increased RDS and RS but decreased SDS (Chung et al. 2009), underscoring that ANN's effect on SDS, like that of HMT, is source‐ and protocol‐dependent rather than a uniform rule. On feasibility, ANN's principal constraint is duration, with individual protocols reported up to 240 h per cycle, considerably longer than typical HMT treatment times.

2.2.3. Comparing HMT and ANN

Both treatments act through a shared underlying transition from disordered to ordered starch domains, governed by moisture content, temperature, and treatment duration (Chi et al. 2019). Within HMT, the only treatment for which starch type comparisons are available, amylose content is the clearest determinant of outcome. High‐amylose starches trend toward RS or combined SDS and RS gain, whereas very low amylose starches trend toward SDS specifically, mirroring the amylose‐related divergence seen for HPT in Section 2.1, though only as a suggestive parallel because the datasets were not designed for direct comparison. Practically, HMT's shorter processing time, 1–24 h against up to 240 h per cycle for ANN, is likely the more decisive feasibility factor between the two treatments for industrial scale‐up, ahead of energy input or equipment cost.

2.3. Dry Heat Treatment (DHT)

DHT heats starch at elevated temperature (110–200°C) with moisture kept below 10%, insufficient to induce gelatinization (Flores‐Silva et al. 2017), distinguishing it from HMT and ANN, which use higher moisture and involve partial or full granule swelling. This limited water plasticizes starch chains, permitting hydrogen‐bond reorganization in amorphous and crystalline regions without collapsing the granule, producing an intermediate structure more ordered than native starch but short of the full crystallinity associated with RS (Bei et al. 2026; Bonto et al. 2021).

This accessibility‐reduction account does not hold uniformly across starches. In normal maize starch, DHT at 140°C (4–20 h) raised SDS from 53.05% to 62.48%, an increase of 8.98 percentage points, alongside falling RDS, consistent with crystalline disruption raising SDS (Zou et al. 2020). In blue highland barley starch, SDS instead declined monotonically with DHT severity alongside falling crystallinity and double‐helix content (Liu, Liu, et al. 2023), the same structural direction as maize but the opposite digestibility outcome. In taro starch, SDS rose under moderate DHT even as crystallinity fell monotonically, attributed to increasing short‐range order and hydrophobicity acting independently of long‐range crystallinity (Hui et al. 2024). Buckwheat starch showed no SDS change at all despite sharp, opposing RDS and RS shifts, indicating DHT can redistribute digestibility entirely through an exchange between RDS and RS that bypasses SDS in some starches (Dias et al. 2026). Comparable SDS gains have also been reported in red adzuki bean and waxy rice starches, whereas wheat starch showed only a marginal SDS increase alongside a rise, rather than fall, in RDS (He, Li, et al. 2025; Xu et al. 2018; Zeng, Ma, et al. 2015).

Excessive treatment fails in different directions depending on starch. In taro, exceeding 150°C/5 h reversed the moderate‐treatment trend entirely, pushing RDS above native level while both SDS and RS fell, a genuine overshoot into net digestibility gain (Hui et al. 2024). In sweet potato, the same kind of threshold instead produced a biphasic SDS response with a clear interior peak (three repeated cycles, 11.97%, or 9 h continuous, 9.46%) before declining as RS reaccumulated (Gou et al. 2019). Treatment mode matters independently of severity: In sweet potato, repeated DHT produced higher SDS than continuous DHT at matched durations, whereas continuous DHT produced higher RS, consistent with continuous treatment affecting the granule surface more strongly and repeated treatment affecting internal crystalline structure more strongly (Gou et al. 2019).

SDS gains under DHT carry costs. Reduced gelatinization enthalpy, seen across buckwheat, taro, and blue highland barley, lowers the energy needed for downstream gelatinization‐dependent processing (Dias et al. 2026; Hui et al. 2024; Liu, Zhang, et al. 2023), though the optimal protocol is application‐specific: Moderate DHT suited taro to high‐viscosity, gel‐forming uses, whereas 150°C/5 h suited it instead to low‐viscosity, rapid‐digestion uses (Hui et al. 2024). In taro, retrogradation tendency rose with DHT intensity, meaning the conditions that raised SDS also increased susceptibility to structural reordering during storage (Hui et al. 2024); in sweet potato, paste clarity fell to a minimum at the same intensity at which SDS peaked, indicating the digestibility optimum and the paste‐stability minimum coincide rather than being independent (Gou et al. 2019). Blue highland barley starch also darkened measurably with DHT severity, consistent with caramelization, affecting sensory acceptability independent of digestibility benefit (Liu, Liu, et al. 2023).

Taken together, DHT's SDS‐enhancing potential depends on a finely balanced degree of structural reorganization, as with HMT and ANN. That balance point is not fixed: the treatment intensity at which SDS peaks, the fraction that dominates beyond it, and the associated stability and quality costs all vary by starch source, indicating no single structural mechanism governs the SDS response to DHT across starch types.

2.4. Ball Milling

Ball milling is a mechanical physical modification technique that disrupts starch granule structure through high‐energy shear, impact, and frictional forces. Unlike hydrothermal treatments, ball milling induces extensive granular fragmentation and particle size reduction, leading to the breakdown of crystalline regions and increased amorphization. These structural alterations significantly influence starch digestibility by modifying enzyme accessibility and subsequent molecular reorganization (Bangar et al. 2023; Han et al. 2022).

Recent studies demonstrate that the effect of ball milling on SDS is highly dependent on the native starch structure and digestibility profile. In starches with a high native RS content, such as potato starch, controlled ball milling can partially disrupt the highly resistant crystalline regions, converting part of the RS fraction into SDS. Pearson correlation analysis further confirmed that changes in the SDS fraction were closely associated with the loss of long‐range crystalline order (Han et al. 2022). For example, the SDS content of potato starch increased from 1.95% to 50.00%, whereas RS decreased from 93.08% to 48.20%, indicating that a substantial proportion of the resistant fraction was transformed into SDS rather than directly into RDS (Lv et al. 2019). Nevertheless, the disruption of the semicrystalline structure and its conversion into a more amorphous state also increased overall enzyme accessibility, making ball‐milled potato starch more digestible than the native starch (Lv et al. 2019). In contrast, starches with relatively low native RS content are more susceptible to excessive structural damage during milling. High milling intensity promotes depolymerization, weakens intermolecular interactions, and destroys the ordered crystalline architecture, thereby accelerating enzymatic hydrolysis. For example, planetary ball milling of wheat starch at 500 rpm for 60 min reduced the SDS content from 12.33% to 1.97%, whereas in B‐type wheat starch granules, SDS decreased from 12.58% to 4.28% after 1 h of milling (Han et al. 2022). These changes were attributed to the conversion of SDS and RS fractions into RDS as a result of extensive mechanical disruption.

Therefore, unlike most physical modification methods that consistently enhance SDS through molecular reorganization, the effect of ball milling is starch‐dependent and reflects the balance between controlled disruption of highly resistant structures and excessive destruction of the crystalline network.

2.5. Microwave Processing

Microwave processing heats starch volumetrically: Dielectric friction from water and polar groups disrupts double helices and crystalline lamellae, with the extent of disruption tied to structurally bound water. B‐type crystallites hold up to 36 inter‐helical water molecules per unit cell versus four in A‐type; consequently, B‐type starches consistently lose more crystallinity than A‐type under equivalent treatment. This disruption enables reorganization on cooling, which can partially recrystallize into structures with reduced enzyme accessibility (Xu et al. 2019).

Microwave treatment increases SDS by partially disrupting granular, crystalline, and double‐helical structures, improving enzyme access to internal binding sites. In normal maize starch and potato starch treated at 30% moisture and 6.63 W/g for 5 min, SDS rose from 28.64% to 47.16% (normal maize starch) and from 23.12% to 42.03% (potato starch), gains of 18.52 and 18.91 percentage points, respectively; in normal maize starch this was accompanied by a 24.25 percentage point fall in RS as both SDS and RDS increased (Xu et al. 2019).

The magnitude of this response depends on granule architecture. Porous A‐type granules, such as normal maize starch, allow enzymes to penetrate through surface channels (“inside‐out” digestion), so microwave‐induced loss of crystallinity and branching directly enhances digestion. B‐type granules, such as potato, have a smooth, rigid surface that initially resists enzyme entry; only at high power (6.63 W/g as reported), once surface hollows and damage form, does enzyme access increase, sharply raising both SDS and RDS at the expense of RS (Xu et al. 2019). Consistent with this, B‐type starches lose more crystallinity than A‐type under equivalent treatment (potato 21% vs. maize 7.8%), whereas C‐type chestnut starch converts toward the A‐type pattern, losing up to approximately 56% crystallinity before partially recovering with longer treatment (Liu, Wen, et al. 2025; Xu et al. 2019).

Granule size adds a further, largely independent factor: small granules (quinoa, amaranth; 1–4 µm) lose crystallinity completely above 75°C, whereas maize starch (approximately 15 µm, same polymorph) shows only a 7% loss and a shift from A‐ to B‐type under the same conditions, reflecting its lower surface area, thicker crystalline lamellae, and longer amylopectin chains (Wei et al. 2025). Conclusions from one starch source should not be generalized without this context.

Treatment time affects the formation of both SDS and RDS as well as the reduction of RS, and further treatment can reverse these gains rather than sustaining them. In chestnut starch, SDS rose from 12.5% in native starch to a peak of 32.3% at 90 s, then fell sharply to 14.1% by 150 s, below even the 60 s level, before rising slightly to 17.3% at 180 s; RS moved inversely throughout, indicating the system had not stabilized even after 180 s (Liu, Wen, et al. 2025).

Microwave's volumetric heating supports continuous processing and largely preserves product appearance, an advantage for scale‐up over autoclaving or boiling (Wei et al. 2025). At higher power and longer duration, however, the dominant mechanism shifts from disrupting supramolecular organization to degrading the starch molecule itself: in maize starch held at 150°C for up to 2 h, amylose content fell by 67% through chain scission and gel viscosity dropped below detection, compromising functionality even where digestibility improved (Wei et al. 2025). Short, moderate‐power treatments therefore preserve starch integrity but give smaller, less predictable SDS gains, whereas prolonged high‐intensity treatment favors digestibility at the cost of functionality, so industrial relevance depends on weighing the SDS outcome against the accompanying loss of pasting and gel properties.

2.6. Cold Plasma Treatment

Cold plasma is a nonthermal technique in which starch is exposed to partially ionized gases containing reactive species (ions, electrons, radicals, and UV photons) under ambient or low‐temperature conditions, enabling surface‐level modification without extensive thermal degradation (Sun, Sun, et al. 2023; Xue et al. 2026); cold plasma increases SDS and RS mainly through two concurrent changes, depolymerization and cross‐linking, both driven by the reactive species generated within the plasma (Okyere et al. 2022). Depolymerization produces low‐molecular‐weight fragments that reassociate into double helices resistant to enzymatic attack (Ge et al. 2021), whereas cross‐linking joins hydroxyl groups from two glucose units via a new ether (C–O–C) bond, forming a more rigid, less accessible chain network (Liu, Zhang, et al. 2025). Reactive oxygen and nitrogen species can further oxidize surface hydroxyl groups, mainly at C‐6, to carbonyl and carboxyl groups that sterically hinder enzyme binding (Wongsagonsup et al. 2026).

The extent of these changes depends on granule structure and botanical origin. Tightly packed crystalline starches, with branching points embedded within the crystalline regions, offer fewer accessible sites and favor cross‐linking. Loosely packed starches, with branching points in the amorphous regions, are more accessible and favor depolymerization and cross‐linking of short chains alongside formation of longer chains, typically producing the greater SDS and RS increase (Liu, Zhang, et al. 2025). Higher amylose content is independently associated with a greater RS response through polymerization of amylose into larger molecules under plasma exposure (Sun, Sun, et al. 2023).

Processing intensity governs the direction of the effect. Short, moderate treatment (approximately 1–21 min), across radiofrequency systems using argon–carbon dioxide or compressed‐air mixtures, in both granular and non‐granular starches and across botanically distinct sources, consistently lowers RDS while raising SDS and RS, with the effect increasing progressively within this range; corn and potato starch treated with radiofrequency air plasma, for example, showed a progressively larger fall in RDS and rise in SDS and RS as duration increased from 3 to 21 min (Ge et al. 2021; Liu, Zhang, et al. 2025). Beyond a threshold, this reverses: dielectric barrier discharge treatment at high voltage (15–20 kV) raises RDS and lowers RS, as harsher conditions damage the granule surface and interior, generating excess short‐chain fragments that expose further enzyme‐binding sites. This mirrors the pattern seen with other physical modification methods, where intensity governs whether the dominant effect is resistance‐conferring reorganization or digestibility‐increasing breakdown (Ge et al. 2021; Liu, Zhang, et al. 2025; Wongsagonsup et al. 2026).

Digestibility gains can also cost functional quality. Comparing gas‐solid plasma (dry starch exposed directly to the discharge) against gas–liquid plasma (starch suspended in water) at matched power and duration in mung bean starch, the gas–liquid configuration produced substantially greater structural disruption (amylose falling to approximately 3.00 g/100 g vs. approximately 13.00 g/100 g), but peak pasting viscosity collapsed from 4265.67 cP to as low as 18.72 cP, leaving almost no typical pasting profile. Conditions tuned to maximize structural disruption, and by extension resistant‐fraction formation, can therefore remove the paste‐forming functionality most food applications require (Xue et al. 2026).

Variability across studies may also reflect differences in treatment delivery rather than intensity alone. Low‐pressure systems offer even, controllable distribution but require vacuum equipment and batch operation, whereas atmospheric systems (dielectric barrier discharge and plasma jets) are cheaper and continuous but distribute reactive species less evenly. Differences in reaction gas, reactor geometry, and delivery medium across studies grouped under the single label “cold plasma treatment” complicate comparison across the literature and pose a practical barrier to scale‐up, alongside the unresolved regulatory status of plasma‐modified starch (Thirumdas et al. 2017).

Commercial translation also faces nontechnical barriers: Safety and regulatory approval remain unresolved for this and other nonthermal technologies, and equipment cost, scale‐up from lab‐scale reactors, and reliance on specific process gases add further constraints (Wongsagonsup et al. 2026).

2.7. Ultrasound Processing

Ultrasound alters starch structure through acoustic cavitation: The formation and collapse of microbubbles generate localized shear, pressure gradients, and thermal effects that disrupt granules and influence molecular mobility (Lu et al. 2018; Shi et al. 2025). Unlike conventional thermal treatments, ultrasound promotes structural disruption and molecular reorganization simultaneously, so the resulting digestibility profile depends on the balance between granule damage and the capacity of disrupted chains to reorganize into enzyme‐restricting ordered structures (Bei et al. 2026).

Ultrasound is more commonly associated with an RS increase than an SDS increase, though SDS gains have been reported under specific conditions, in select starches, or in combination with other modification strategies. Moderate treatment has been linked to increased structural ordering and SDS gains of roughly 5%–15% depending on conditions (Hao et al. 2025), but the response is governed less by intensity alone than by starch botanical origin, crystalline polymorphism, and amylose content. Under matched conditions (28 kHz), ultrasonic treatment of native corn (A‐type) and potato (B‐type) starch primarily raised RS while lowering SDS, whereas C‐type pea starch showed a modest SDS increase, demonstrating strong polymorphism dependence (Xiao et al. 2021). The same study found sonication temperature modified this further: Low temperatures promoted molecular reorganization and higher relative crystallinity in A‐type corn starch, whereas B‐type potato and C‐type pea starch instead underwent greater crystalline disruption, yielding higher digestibility and lower RS retention. Native crystalline architecture therefore determines whether cavitation predominantly drives reorganization or degradation under otherwise comparable conditions. Combining ultrasound with other strategies, such as starch‐lipid or starch‐protein complexation, has produced larger SDS gains than ultrasound alone, suggesting it may function better as a facilitating technology than a standalone SDS‐maximizing approach (Rostamabadi et al. 2024).

The SDS response is also highly dependent on treatment intensity and duration. Excessive sonication induces chain scission, increased porosity, and crystalline disruption, generally decreasing SDS and increasing RDS (Hao et al. 2025; Lu et al. 2018), though processing parameters (frequency, duration, starch concentration, probe vs. bath configuration, effective cavitation energy) all affect outcomes independently of intensity, complicating comparison across studies (Hao et al. 2025; Lu et al. 2018; Shi et al. 2025). Applied power level drives a further duality: Low power (100–200 W) primarily causes surface erosion that facilitates enzyme penetration, whereas higher power (300–600 W) can generate enough thermal energy to reorganize degraded chains into stable V‐type crystallites on cooling (Rostamabadi et al. 2024).

Industrially, ultrasound offers low processing temperatures, short treatment times, reduced chemical usage, and environmentally favorable processing, but optimization is challenging. Conditions favoring SDS formation do not necessarily maximize RS or preserve functional properties such as viscosity, swelling, and gel formation, and starch‐source variability is considerable (Wongsagonsup et al. 2026). Overall, ultrasound is better characterized as a versatile processing platform than a universally effective SDS‐promoting technology, its success depends on carefully optimized, starch‐specific conditions, and combined modification approaches appear more capable of producing stable SDS‐rich starches than ultrasound alone.

2.8. Controlled Heating–Cooling Treatments

Controlled heating–cooling treatment is a widely recognized strategy for enhancing SDS. Starch is first gelatinized by heating, then subjected to controlled cooling, storage, or freeze–thaw, promoting reassociation of starch chains, particularly amylose, into more ordered structures via retrogradation without chemical reagents (Raghunathan et al. 2025; Tian et al. 2013; Yang et al. 2025; Zhang et al. 2018). This amorphous‐to‐semicrystalline transition reduces enzyme accessibility and slows hydrolysis, but the outcome depends on the balance between chain mobility and structural stabilization rather than retrogradation alone, with storage temperature particularly influential because it governs nucleation and crystal growth rates. The greatest SDS gains occur under conditions promoting partial reassociation without excessive crystallization. Temperature‐cycled storage produced higher SDS than isothermal storage in waxy potato starch (38.63%). In Tartary buckwheat starch, a single temperature‐cycling treatment (4/25°C) produced the highest SDS (35.25%), while increasing cycle number or isothermal storage at 25°C favored RS instead (Wu et al. 2021); similarly, triple retrogradation reduced SDS relative to dual retrogradation as continued rearrangement generated increasingly stable, less digestible crystalline structures (Luong et al. 2024; Raghunathan et al. 2025). The conditions that maximize SDS are therefore not those that maximize RS: SDS is favored by intermediate structural ordering, whereas extensive recrystallization shifts the profile toward RS3.

Mechanistically, tightly packed double helices and semicrystalline regions impose steric hindrance that restricts enzyme penetration and α‐amylase diffusion, often via a transition from A‐type crystallinity toward B‐ and V‐type structures (Wu et al. 2021), with dense, high‐fractal‐dimension aggregates providing a further physical barrier (Zheng et al. 2022). Greater structural order does not, however, invariably raise SDS: Once reassociation reaches highly stable crystalline domains, the profile favors RS instead.

Freezing is a more intensive form of cooling, in which ice‐crystal formation simultaneously promotes molecular reassociation and mechanically disrupts granules, producing a less uniform response than refrigerated storage, dependent on freezing temperature, rate, cycle number, starch source, and food matrix. Spiral tunnel and liquid nitrogen spray freezing raised SDS in starch isolated from unfermented wheat dough (34.43%–37.27% and 36.67%, respectively) (Yang, Zheng, et al. 2021) whereas repeated freeze–thaw of Chinese yam starch raised SDS only modestly (0.78%–2.50%) while driving extensive RS formation (>96.7%), indicating structural stabilization outpaced slowly digestible domain formation (Huang et al. 2025). Freezing therefore does not consistently enhance SDS; the outcome depends on whether ice‐crystal disruption or molecular reassociation dominates under the chosen conditions.

Overall, controlled heating–cooling is among the most effective physical approaches for raising SDS, but the outcome depends on balancing gelatinization, chain mobility, and subsequent ordering. Partially ordered structures maximize SDS, whereas excessive crystallization favors RS; optimization should therefore target slowly digestible domains rather than simply maximizing retrogradation.

2.9. Extrusion

Extrusion is among the most variable physical processing methods for starch digestibility, capable of reducing or enhancing SDS depending on the balance between gelatinization, molecular degradation, and subsequent reorganization during cooling (Huang et al. 2022; Roman et al. 2019). Unlike most physical modifications, extrusion applies heat, shear, pressure, and moisture simultaneously, driving an order–disorder–reorder transition; the digestibility outcome therefore depends on whether disrupted chains reorganize before enzymatic digestion, governed collectively by feed moisture, barrel temperature, screw speed, specific mechanical energy (SME), starch composition, and post‐extrusion cooling. Feed moisture and SME appear most influential, as they determine the balance between fragmentation and chain reassociation (Huang et al. 2022).

Severe extrusion generally decreases SDS, whereas moderate extrusion, or extrusion followed by controlled reassembly, can increase it. Under severe thermomechanical conditions, extrusion disrupts crystallinity, destroys double‐helical order, and fragments amylopectin, increasing enzyme accessibility. Extrusion of pure corn, potato, and pea starches at high moisture (42%) across a range of screw speeds (80–320 rpm) reduced RS most sharply, falling to roughly a quarter to a third of native levels across all three sources, whereas SDS declined more modestly and unevenly by source, alongside falling crystallinity, gelatinization enthalpy, and peak viscosity in all three starches (Wang, Li, et al. 2023). A parallel effect was seen with moisture content itself as the extrusion variable: Pure high‐amylose jackfruit seed starch extruded at reduced moisture (10% vs. 20%) lowered SDS and RS from 24.08% and 25.99%–17.59% and 16.60%, respectively, with the authors attributing the more digestible product at low moisture to lower crystallinity and weaker short‐range order in the extrudate, both of which rose with increasing moisture (Zhang et al. 2022).

In contrast, moderate conditions favoring partial disruption and reorganization raised SDS. Pure high‐amylose maize starch extruded across a range of screw speeds (250–600 rpm), barrel temperatures (100–160°C), and feed moisture (16.4%–22.5%) rose in SDS from 20.4% native to 27.5%, reversing the trend seen in native starches where higher amylose content is normally associated with lower SDS, whereas normal maize starch under the same conditions retained a smaller SDS fraction of up to 15.6% relative to its native 27.8%, and waxy maize starch became almost entirely rapidly digestible (Robin et al. 2016). The SDS gain in extruded starch was attributed to a combination of surviving native crystalline structures, retrograded starch formed on cooling, and amylose‐lipid complexes generated from the starch's own lipid content during extrusion, the latter supported by a new endothermic transition near 130°C appearing after extrusion that was absent in the native starch, though the authors present these as contributing possibilities rather than a fully resolved single mechanism (Robin et al. 2016).

Taken together, these studies indicate that SDS formation under extrusion is not a fixed outcome of the technology but a narrow, source‐dependent window between two failure modes. Too much structural disruption, from high moisture, high shear, or high temperature, converts starch to RDS by destroying crystalline and double‐helical order faster than any new order can form. Too little disruption leaves the native structure largely intact and offers no fragmented chains to reassemble. SDS gains occurred only in the space between these extremes, where disruption was sufficient to loosen the granule but incomplete enough to leave chain segments free to recrystallize, complex with lipids, or retain partial order on cooling. Extrusion is therefore best viewed as a technology whose SDS outcome depends on balancing the degree of structural disruption against the extent of subsequent molecular reassembly, rather than on processing intensity alone (Zhao et al. 2024; Bhattarai et al. 2025).

2.10. Magnetic Field Treatment

Magnetic field treatment is an emerging nonthermal physical modification technology recognized as a green and energy‐efficient approach for modifying starch structure without the use of chemical reagents (Jiang et al. 2025). Static and alternating magnetic fields modify molecular interactions within starch granules by promoting chain rearrangement, strengthening hydrogen bonding, and enhancing short‐range molecular order, thereby reducing enzyme accessibility and slowing starch hydrolysis (He, Du, et al. 2025). Although the precise molecular mechanism has not been fully elucidated, studies across different starch systems consistently indicate that magnetic field treatment promotes structural ordering while preserving the granular architecture when appropriate processing conditions are applied (Niu et al. 2025; Song et al. 2026).

Current evidence demonstrates that magnetic field treatment can increase SDS in native starches, although the response is highly dependent on starch source and processing conditions and the magnitude of improvement is modest and strongly dependent on processing intensity (Singh et al. 2025). In germinated corn starch, a static magnetic field of 30 mT applied for 2 h increased SDS by 27.71% and RS by 9.43%, whereas RDS decreased by 8.83% (Jiang et al. 2025). Similarly, normal maize starch exposed to a low‐intensity alternating magnetic field (2 mT) exhibited reduced RDS together with increased SDS and RS fractions, whereas higher magnetic field intensities (4–10 mT) progressively diminished these benefits, demonstrating that magnetic field intensity is a critical determinant of digestibility behavior (Niu et al. 2025).

Comparison of the available studies further indicates that starch response depends on the combined effects of magnetic field intensity, molecular ordering, and structural disruption. Low magnetic field intensities promote molecular rearrangement and strengthen ordered regions that restrict enzymatic accessibility, whereas excessive field intensity increases surface porosity and internal fissures, facilitating enzyme penetration and partially reversing the SDS‐promoting effect (Niu et al. 2025). This nonlinear response is consistent with the pattern seen across most other physical modification techniques in this review, where an optimum treatment window rather than a linear dose‐response relationship governs SDS formation, and it reinforces the need to identify that optimum for each starch source and processing method individually.

From an industrial perspective, magnetic field treatment offers several advantages, including noncontact processing, low energy consumption, and the absence of chemical additives. However, current evidence is largely restricted to laboratory‐scale studies, with considerable variation in magnetic field strength, treatment duration, moisture content, and magnetic field configuration, making direct comparison between studies difficult. Furthermore, little information is available regarding the stability of the induced SDS during storage or subsequent food processing. Therefore, although magnetic field treatment shows promise as a green technology for enhancing SDS in native starches, further research is required to optimize processing conditions, validate the technology across a wider range of botanical starch sources, and establish its industrial feasibility and long‐term effects on starch digestibility (Niu et al. 2025; Song et al. 2026).

The examples presented in Table 1 represent selected case studies of physical modification methods that enhance SDS, chosen based on the availability of quantitative SDS data before and after treatment, together with clearly defined processing conditions and associated structural characterization. Collectively, these studies demonstrate the wide variability in SDS response across different starch sources and modification approaches. Treatments that induce substantial structural reorganization generally produce greater increases in SDS, whereas milder treatments result in more limited effects.

As illustrated in Table 1, the extent of SDS enhancement varies considerably, ranging from minor increases, such as DHT of maize starch (ΔSDS ∼ 9%), to substantial changes, such as ball milling of potato starch (>48% increase). Intermediate responses are also observed, for example, in microwave‐treated maize starch (∼18% increase). These differences highlight the strong influence of both starch botanical origin and treatment severity on digestibility outcomes.

2.11. Stability and Recovery of SDS: An Underexplored Question

Two questions remain largely unaddressed by the physical modification literature reviewed in Sections 2.1–2.10: Whether treatment can be deliberately staged to avoid crossing the severity threshold at which SDS gains reverse, and whether SDS, once formed, persists afterward. Neither has been the direct object of study; the literature was not designed to answer them, and what follows is inferred from studies whose primary aims lay elsewhere.

On the second question, evidence is almost entirely absent because the majority of the previous studies typically measure digestibility once, immediately after treatment, rather than tracking it over time. Consequently, the stability of SDS during reheating, extended storage, or subsequent processing remains largely unknown across the physical modification literature.

A small number of incidental findings hint at what deliberate control might look like, without constituting evidence that such control has been demonstrated. Extruded starch in which SDS was substantially reduced during processing was later found, in the same study, to have increased up to 3.5‐fold on subsequent storage (Liu et al. 2021), suggesting that a posttreatment SDS value need not be the final outcome, though this was reported as an observation rather than pursued as a recovery method. One plausible direction suggested by such findings is dual or sequential modification, combining an initial disruptive treatment with a second step that promotes controlled reorganization, which is examined in detail in the following section.

3. Synergistic Effects of Dual Physical Modifications

Sequential physical processing treatments can induce complex structural transformations within starch granules because each treatment modifies different aspects of the granular and molecular architecture. In such systems, the first treatment typically alters granule integrity or surface properties, thereby increasing the susceptibility of the starch matrix to further structural rearrangement during the subsequent processing step (Babu et al. 2019; Yang et al. 2025). For example, dry heating pretreatment decreases molecular weight by primarily degrading α‐1,6‐glycosidic linkages, increasing the flexibility of starch chains and allowing them to be more effectively reorganized into highly ordered structures during subsequent annealing (Chi et al. 2019).

The studies summarized in Table 2 demonstrate that some dual physical treatments can substantially modify starch digestibility and, under optimized conditions, significantly enhance SDS formation because the first treatment creates a starch matrix that is more responsive to the second processing step. However, the magnitude of SDS enhancement varies considerably depending on starch botanical origin, treatment sequence, and processing intensity. For example, the combination of cold plasma and microwave treatment increased SDS in lotus root starch from 8.17% to 36.95%. Cold plasma modifies the granule surface through oxidative etching, increasing surface roughness and permeability, whereas the subsequent microwave treatment promotes more extensive molecular rearrangement within the disrupted granules (Sun, Sun, et al. 2023). Compared with either treatment alone, the combined process produced greater reductions in crystallinity and more pronounced changes in granule morphology, resulting in a substantial conversion of RDS into SDS (He, Li, et al. 2025; Sun, Sun, et al. 2023).

TABLE 2.

Examples of dual physical processing strategies and their effects on slowly digestible starch (SDS) formation.

Starch source	Dual treatment	SDS before (%) a	SDS after (%) a	Absolute change in SDS (%) b	Structural observations	Reference
Sweet potato	Repeated dry heat, 130°C, 3 h/cycle, cooled to 25°C for 60 min, 3 cycles	5.05	11.97	+6.92	Early cycling disrupted crystalline order without complete resistant reordering	Gou et al. (2019)
Wheat	Repeated heat–moisture treatment (30% moisture; 120°C, 2 h/cycle; 5 cycles)	21.87	48.36	+26.49	Repeated heating/cooling promoted molecular rearrangement and water redistribution	Su et al. (2020)
Lotus root	Microwave + cold plasma (300 W + 60 s CP, 40% moisture)	8.17	36.95	+28.78	Microwave disruption followed by plasma‐assisted molecular rearrangement increased SDS	Sun, Sun, et al. (2023)
Rice (high amylose)	Dual retrogradation (boiling → 4°C → boiling → 4°C)	39.30	56.70	+17.40	Ordered double‐helical structures restricted enzyme access	Tian et al. (2013)
Foxtail millet	Ultrasonication followed by annealing	11.42	13.84	+2.42	Ultrasonication fragmented chains; annealing promoted crystalline perfection	Babu et al. (2019)
Quinoa	Microwave + heat–moisture treatment (25% moisture; 4.8 W/g; 90 s)	16.73	25.59	+8.86	Rearrangement into more ordered crystalline structure	Liu, Zhang, et al. (2023)
Maize	Dry heating + annealing	2.08	9.98	+7.90	Flexible chains rearranged into highly ordered lamellae	Chi et al. (2019)
Waxy rice	Annealing + heat–moisture treatment	45.5	54.6	+9.1	Decreased relative crystallinity and molecular weight	Zeng, Ma, et al. (2015)
Corn	High‐pressure + annealing (300 MPa, 30 min, 70% moisture)	15.51	19.18	+3.67	Pressure‐induced molecular rearrangement increased ordering	Ye et al. (2026)
Red adzuki bean	Repeated annealing (55°C, 96 h; 1:3 w/v)	7.39	22.64	+15.25	Increased relative crystallinity and molecular reorganization	Xu et al. (2018)

Note: Comparisons across studies should be interpreted cautiously because digestion protocols, sample preparation, and starch fractionation methods may differ.

^a SDS values are reported as presented in the cited studies.

^b Absolute change in SDS was calculated as SDS after–SDS before (%).

Other sequential treatments operate primarily through mechanisms associated with controlled recrystallization. Microwave heating followed by storage promotes retrogradation of starch chains during cooling, resulting in the formation of ordered crystalline domains that reduce enzymatic accessibility (Liu, Liu, et al. 2023; Xu et al. 2019). Similarly, combined microwave‐assisted HMT increased SDS in quinoa starch from 16.73% to 25.59% under an optimum moisture content of 25%, demonstrating that careful control of moisture content is essential to promote molecular rearrangement without excessive gelatinization (Liu, Liu, et al. 2023). Repeated HMT also facilitates progressive molecular reorganization through multiple thermal cycles, enabling amylopectin chains to gradually rearrange into more stable crystalline structures (Brahma and Sit 2020; Huang et al. 2015). Likewise, repeated ANN increased SDS in red adzuki bean starch from 7.39% to 22.64% by improving crystalline perfection and strengthening intermolecular interactions (Wu et al. 2021). Furthermore, sequential dry heating (130°C) followed by annealing (50°C) increased SDS in normal maize starch from 2.08% to 9.98%, confirming that preliminary molecular disruption can enhance subsequent structural ordering during hydrothermal treatment (Chi et al. 2019).

The effectiveness of dual physical modification therefore arises from the interaction of two complementary structural processes: initial disruption or activation of the starch granule followed by enhanced molecular rearrangement, chain reassociation, or recrystallization (Babu et al. 2019). Nevertheless, these synergistic effects are not universal. Although some combinations preferentially increase SDS, others promote RS formation or reduce SDS, depending on starch crystalline type, amylose content, treatment sequence, and processing severity. This highlights that combining two physical treatments does not necessarily produce additive improvements in SDS but rather requires complementary mechanisms that promote controlled structural reorganization while preserving sufficient molecular order. In some instances, these sequential treatments can also produce thermally stable SDS that resists disruption during cooking, as demonstrated in maize starch where HMT followed by sonication rearranged double helices within the crystalline lamella into more stable crystallites (Flores‐Silva et al. 2017). The available evidence indicates that the success of dual physical modification depends not only on the combination of technologies employed but also on optimizing treatment conditions for individual starch sources, which remains an important consideration for future industrial application.

4. Enzymatic Modification of Starch Structure

Enzymatic modification regulates starch digestibility by acting directly on the molecular architecture of amylose and amylopectin, where chain‐length distribution and branching density govern the rate‐limiting steps of hydrolysis (Li, Zhang, et al. 2022; Miao and BeMiller 2023). By selectively cleaving or rearranging α‐1,4‐ and α‐1,6‐glycosidic linkages, these treatments alter how glucan chains reassociate into ordered structures during cooling or storage, thereby changing enzyme accessibility (Freitas et al. 2025; Huang et al. 2025). For instance, sequential treatment with maltogenic α‐amylase and branching enzyme increased the degree of branching in sweet potato starch from 12.02% to 44.06%, while reducing average chain length from 18.02 to 12.32 degree of polymerization (DP), shifting digestion toward slower hydrolysis and greater SDS (Bello‐Perez et al. 2020; Wang, Mi, et al. 2023).

This shift arises primarily from structural barriers to enzyme–substrate binding. Productive binding by pancreatic α‐amylase requires a sequence of five adjacent, linearly aligned glucose units; enzymatic treatments that raise branching density or shorten chains below this threshold weaken digestion continuity (Bello‐Perez et al. 2020; Li et al. 2026). The resulting highly branched α‐limit dextrins are further hydrolyzed more slowly, as mucosal α‐glucosidases cleave α‐1,6 linkages much less efficiently than α‐1,4 bonds (Korompokis et al. 2021). SDS content appears to be favored by amylopectin chain‐length distributions dominated by either very short A‐chains (DP < 13) or elongated B‐chains, which may promote complex crystalline arrangements or steric hindrance that limits enzymatic accessibility (Huang et al. 2025); higher branching density also restricts glucan chain mobility after gelatinization, extending the time needed to reform ordered structures (Wang, Mi, et al. 2023). SDS formation is therefore governed by the combined effects of chain‐length distribution, branching architecture, and molecular reorganization, rather than any single parameter, meaning different enzymatic strategies can enhance SDS through distinct, complementary structural pathways (Miao and BeMiller 2023).

Direct SDS quantification is not reported uniformly across recent enzymatic studies: some report explicit SDS values before and after treatment, whereas others report only RS, estimated glycemic index, or general hydrolysis behavior. Studies of the latter kind are treated here as mechanistic support only, unless SDS is directly quantified using an established fractionation method, as differences in analytical method may themselves contribute to the variability reported across the literature.

4.1. Debranching Enzymes

Debranching enzymes, particularly pullulanase and isoamylase, act by hydrolyzing α‐1,6‐glycosidic linkages in amylopectin. This reduces branching density, generating more linear glucan chains capable of reassociating on cooling into double helices and ordered crystalline domains. Where this restructuring is appropriately controlled, enzyme accessibility decreases and SDS increases; where debranching is excessive; however, the system shifts further toward RS‐rich structures rather than maximizing SDS (Lee et al. 2010; Li, Wang, et al. 2022; Miao and BeMiller 2023). The extent of debranching is therefore a critical determinant of whether starch develops predominantly slowly digestible or highly resistant domains, rather than debranching being uniformly beneficial to SDS.

Quantitative evidence supports this across several botanical sources. Rice starch treated with pullulanase raised SDS from 7.21% to 30.73%, while combining β‐amylase with debranching enzymes yielded a substantially lower SDS content of 18.13%, indicating that enzyme combinations can be used to tailor the digestibility outcome rather than simply maximize it (Hutabarat and Stevensen 2023). Controlled α‐amylase treatment of potato starch similarly raised SDS from 18.3% to 28.1% by restructuring amorphous and crystalline regions (Hutabarat and Stevensen 2023).

Controlled pullulanase debranching provides direct, quantitative evidence for this effect. In waxy corn starch, increasing pullulanase debranching degree raised SDS content from 2% to 23%, alongside a rise in RS from 13% to 19% and a fall in weight‐average molecular weight of starch from 450 to 100 kDa, consistent with debranching generating shorter, more linear chains capable of partial reassociation into ordered, slowly digested structures (Liu et al. 2017). Posttreatment structuring can raise this further: sequentially fractionating pullulanase‐debranched waxy maize starch with ethanol, narrowing the chain‐length distribution in a stepwise manner, raised SDS to 30.0% and RS to 58.6%, the highest values obtained among the fractions tested, attributed to more homogeneous, densely packed crystallites formed from the narrowed chain population (Zhang et al. 2024).

These studies indicate that SDS formation depends on achieving an optimal degree of debranching rather than maximizing branch removal. Selective cleavage of α‐1,6 linkages generates linear chains capable of controlled reassociation into ordered, slowly hydrolyzed structures, whereas excessive debranching drives extensive recrystallization and favors RS instead. Enzyme dosage, reaction time, and posttreatment storage conditions are therefore equally important in determining the final digestibility profile (Hutabarat and Stevensen 2023; Liu et al. 2017).

4.2. Maltogenic α‐Amylase

Maltogenic α‐amylase primarily hydrolyses α‐1,4‐glycosidic linkages, often acting near nonreducing chain ends, and can also promote transglycosylation. Its effect on starch digestibility is therefore not a simple depolymerization process; instead, it reshapes amylopectin fine structure by shortening longer chains and increasing the relative abundance of shorter and intermediate chains that may subsequently reorganize into more ordered arrangements. Under controlled conditions, this restructuring can increase SDS, whereas excessive hydrolysis may generate smaller, more rapidly digestible fragments (Hutabarat and Stevensen 2023; Wang, Mi, et al. 2023).

Representative studies demonstrate the effectiveness of this approach across different starch systems. It has been reported that the maximum SDS content reached 38.5% when pea starch was treated with 60 ppm maltogenic α‐amylase for 4 h at 60°C, which was attributed to modifications in chain‐length distribution and the formation of branching structures favorable for slower digestion. The same study further showed that, in pulse and maize starches, maltogenic α‐amylase shortened amylose chains and longer amylopectin branches without altering the overall WAXD pattern. Despite minimal changes in crystal type, these structural modifications increased the proportions of SDS and RS, indicating that digestibility can be effectively reduced through molecular remodeling rather than extensive changes in crystalline structure (Li, Fu, et al. 2021; Zheng et al. 2022).

This pattern, however, depends heavily on reaction conditions. A direct comparison on the same substrate, pea starch, illustrates this: a maximum SDS content of 38.5% using a brief, low‐dose treatment (60 ppm maltogenic α‐amylase, 4 h, 60°C) (Zheng et al. 2022), whereas treating pea starch with a substantially higher enzyme loading (1% w/w, equivalent to 110 MANU/g starch) for up to 24 h produced no significant change in SDS (3.1% control vs. 2.0% treated), a pattern consistent across all four starches tested in the same study (lentil, faba bean, pea, and normal maize), none of which showed a statistically significant SDS shift despite substantial absolute changes in some cases (Li, Li, et al. 2021). RS, by contrast, increased significantly in all four starches (+5.9, +6.5, +4.2, and +4.7 percentage points, respectively), attributed not to increased branching, a parallel experiment on waxy maize starch found that MGA treatment raised branching yet reduced RS, ruling that out, but to shortened amylose chains regaining sufficient mobility to reassociate into double‐helical crystallites during the digestibility assay's own incubation step. The digestibility shift in the high‐dose, long‐duration treatment instead appeared as RS rather than SDS (Li, Li, et al. 2021).

Enzyme dose and treatment duration, therefore, appear to govern not only the magnitude of the SDS response to maltogenic α‐amylase but also whether an SDS gain is produced at all, with brief, low‐dose treatment favoring SDS formation and prolonged, high‐dose treatment of the same substrate failing to raise SDS and instead favoring RS. Collectively, these studies indicate that the effectiveness of maltogenic α‐amylase depends on achieving sufficient chain remodeling to promote molecular reassociation without excessive depolymerization. Furthermore, its transglycosylation activity can contribute to local chain rearrangement, enabling the formation of more compact glucan conformations that enhance structural stability without extensive depolymerization (Gourineni et al. 2017; Korompokis et al. 2021).

4.3. Β‐Amylase

β‐Amylase is an exo‐acting enzyme that hydrolyzes α‐1,4‐glycosidic linkages from the nonreducing ends of starch chains, releasing maltose units in a stepwise manner. Unlike endo‐acting enzymes, their action is more controlled and does not extensively disrupt the overall granule architecture. Instead, β‐amylase preferentially trims outer chains of amylopectin and amylose, leading to a redistribution of chain lengths and an enrichment of shorter linear segments. This selective hydrolysis promotes conditions favorable for molecular reassociation and structural ordering, particularly during subsequent cooling or storage (Ao et al. 2007; Bello‐Perez et al. 2020).

Under controlled conditions, β‐amylase treatment has been shown to enhance SDS formation by facilitating the alignment and packing of residual chains into double‐helical and crystalline structures that are less accessible to digestive enzymes. The removal of external chains exposes internal regions that can more readily participate in retrogradation, thereby increasing structural density and reducing enzymatic susceptibility (Chi et al. 2022).

Although quantitative SDS data remain limited for β‐amylase‐treated starches, available studies indicate that moderate enzyme treatment, particularly when combined with cooling or retrogradation, promotes the formation of short linear chains that subsequently reassociate into more ordered structures. This effect is commonly associated with an increase in short‐chain amylose fragments and reorganized amylopectin structures that favor tighter packing and reduced enzyme diffusion. Therefore, the primary mechanism by which β‐amylase enhances SDS is through controlled exo‐hydrolysis that generates linear chains and facilitates their subsequent reassociation into ordered, enzyme‐resistant structures (Huang et al. 2025). This selective trimming can generate intermediate chain populations that are particularly favorable for controlled reassociation, contributing to the formation of slowly digestible structures under in vitro conditions (Korompokis et al. 2021).

4.4. Branching Enzymes

Branching enzymes and starch‐active glucanotransferases modify digestibility by introducing or rearranging α‐1,6 linkages, thereby increasing branching density or generating new branched glucan architectures that are less susceptible to digestive enzymes. Unlike debranching enzymes, which simplify amylopectin architecture, these enzymes increase structural complexity and can slow digestion by creating more steric hindrance and altering the packing and accessibility of glucan chains (Chen et al. 2023; Yoon et al. 2017). Consequently, branch introduction represents a fundamentally different strategy for SDS formation, relying on increased molecular complexity rather than chain linearization (Korompokis et al. 2021).

Representative studies demonstrate the effectiveness of this approach. Corn starch treated sequentially with immobilized α‐amylase for 4 h followed by branching enzyme for 6 h exhibited significantly improved slow‐digestion properties together with a higher degree of branching, an increased proportion of short, branched chains, and altered x‐ray diffraction characteristics. These findings indicate that partial hydrolysis followed by controlled branch introduction effectively redirects starch architecture toward slower digestion (Li, Zhang, et al. 2022; Sacco et al. 2026; Yang, Zhao, et al. 2021). These studies demonstrate that increasing branching density alone is insufficient to maximize SDS formation. Rather, enzyme specificity, the extent of branch introduction and substrate characteristics determine whether newly introduced α‐1,6 linkages generate partially ordered structures associated with SDS or interfere with the molecular organization required for slow digestion. Consequently, an optimum degree of branching, rather than maximum branching density, appears to be the key determinant of successful SDS formation.

4.5. Enzymatic Rearrangement and Structural Engineering of Starch

Beyond hydrolytic and branching modifications, emerging enzymatic strategies focus on restructuring glucan architecture without extensive depolymerization. Enzymes, such as amylomaltase and glucanotransferases, catalyze intramolecular and intermolecular transfer reactions, generating reorganized chain architectures that favor the formation of compact and ordered domains, enhancing SDS by promoting structural configurations that resist rapid hydrolysis. This represents a shift from degradation‐driven modification, which relies on altering the DP, to structure‐directed engineering, in which enzyme specificity and reaction conditions instead tailor chain organization and supramolecular assembly, offering an additional pathway for controlling starch digestibility (Korompokis et al. 2021; Li et al. 2019; Zhong et al. 2022).

4.6. Structural Principles Governing SDS Formation under Enzymatic Treatment

Despite the different catalytic modes of debranching enzymes, maltogenic α‐amylase, and branching or transglycosylating enzymes, their effects on SDS converge on a common principle: SDS increases when enzymatic treatment generates a chain architecture that supports partial ordering and reduced enzyme accessibility without causing excessive hydrolysis or over‐stabilized RS‐type crystallization. In practice, this often involves producing intermediate chain populations, modifying branching density, and promoting reassociation into structures that are more slowly digested but not completely resistant. However, the relative contribution of these mechanisms varies among starch botanical sources and enzyme systems, explaining why similar enzymatic treatments may produce markedly different digestibility profiles. Consequently, successful SDS formation depends on optimizing the extent of structural modification rather than maximizing enzyme activity or reaction severity. These observations indicate that enzymatic modification governs not only molecular composition but also the organization of starch into structures with reduced hydrolysis rates, reinforcing the importance of controlled restructuring rather than extensive degradation (Korompokis et al. 2021).

Overall, the evidence positions enzymatic modification as a precision restructuring strategy for SDS design, not a depolymerization approach. Regardless of the enzyme employed, the greatest improvements in SDS are consistently achieved when treatments generate an optimal balance between chain remodeling, molecular reassociation, and partial structural ordering while avoiding excessive fragmentation or extensive RS formation. Therefore, enzyme selection, dosage, reaction time, substrate characteristics, and posttreatment storage conditions should be considered collectively when designing starches with targeted digestibility profiles (Chi et al. 2022). The principal enzymatic restructuring pathways are illustrated in Figure 2, and representative quantitative outcomes are summarized in Table 3.

FIGURE 2.

Enzymatic restructuring of starch and its role in the formation of SDS.

TABLE 3.

Effects of enzymatic restructuring on slowly digestible starch (SDS) formation.

Starch source	Enzyme treatment	Experimental conditions	SDS before (%) a	SDS after (%) a	Absolute change in SDS (%) b	Experimental mechanism	Structural observations	Reference
Pea	Maltogenic α‐amylase	8% (w/v) starch slurry; 60.0 ppm enzyme; 60°C; pH 4.8; 4.0 h reaction time	12.50	38.50	+26.00	The enzyme selectively hydrolyzes α‐1,4‐linkages while facilitating the transfer of glucosyl residues to generate new, digestion‐resistant α‐1,6‐branch linkages	Increased double‐helix stability Enriched short chains (DP < 13; ∼35.78%), concomitantly reducing paste viscosity and freeze–thaw stability	Zheng et al. (2022)
Potato	4,6‐α‐Glucosyltransferase (StGtfB)	5% (w/v) starch slurry; 5 U/g starch; 40°C; pH 6.0; 2.0 h reaction time	34.29	53.22	+18.93	The enzyme cleaves α‐1,4‐glucosidic linkages and transfers glucosyl residues to form new α‐1,6‐glycosidic branch points, resistant to rapid digestion	Molecular restructuring Promoted short branch chains (DP < 13; 77.11%) alongside a reduction in Mw to 4.39 × 106 g/mol	Li, Fu, et al. (2021)
Pea	Pullulanase	5% (w/v) starch slurry; 80°C/10 min precooking; 40 NPUN/g enzyme; 50°C; pH 5.2; 24.0 h reaction time	5.50	17.30	+11.80	Pullulanase specifically cleaves α‐1,6‐glycosidic branch points in amylopectin to liberate linear glucan chains that subsequently retrograde into digestion‐resistant semicrystalline structures	Debranching and recrystallization Elevated linear glucan content (from 27.9% to 87.5%), accelerating retrograded amylose formation	Lu et al. (2018)
Corn	4‐α‐Glucanotransferase (4αGT)	8% (w/v) starch slurry; 10 U/g starch; 75°C; pH 7.5; 4.0 h reaction time	9.40	20.92	+11.52	The enzyme mediates intra‐ and intermolecular transglycosylation to cleave α‐1,4‐linkages and redistribute glucosyl segments, synthesizing novel amylopectin clusters that hinder enzymatic accessibility	Modified chain distribution Suppressed amylose content to 27.2%, with a simultaneous enrichment of short (DP < 13) and long (DP > 30) amylopectin branches	Jiang et al. (2014)
Wheat	1,4‐α‐Glucan branching enzyme	5% (w/v) starch slurry; 0.5 U/mg substrate; 25°C; pH 5.5; 10.0 min reaction time	2.10	17.39	+15.29	Converts α‐1,4‐glycosidic bonds into α‐1,6‐branching at C‐6 positions, enhancing steric hindrance to effectively inhibit amylolytic degradation	Increased short‐chain branching and reduced retrogradation Downregulated amylopectin Mw, while expanding the short‐chain population (DP 6–12) to ∼43.52%	Li, Fei, et al. (2020)
Corn	1,4‐α‐Glucan branching enzyme	20% (w/v) starch; 25 U/g GBE (Stage 1) + 25 U/g GBE (Stage 2); 55°C; pH 7.5; 10 + 10 h (intermediate 30‐min boiling)	12.09	26.88	+14.79	Two‐stage modification creates dense cluster structures and cyclic glucans, enhancing steric hindrance to effectively inhibit amylolytic attack	Formation of clustered structure and short‐chain enrichment Maximized α‐1,6 density (+75.4%) and short chains (33.88%), yielding a 98.9% reduction in amylopectin Mw	Zhu et al. (2025)
Potato	1,4‐α‐d‐Glucan branching enzyme	30% (w/v) starch slurry; 200 U/g enzyme; 50°C; pH 7.5; 10.0 h reaction time	12.70	18.60	+5.90	GBE converts α‐1,4 bonds into new α‐1,6‐branching points, increasing molecular density to effectively inhibit amylolytic access	Increased α‐1,6‐branching and reduced crystallinity Lowered relative crystallinity (11.77%) and chain length, while increasing the short A‐chains (24.82%)	Li et al. (2018)
Potato	Aquabacterium branching enzyme (AqGBE)	6% (w/v) starch; 0.8 mg/g enzyme; 40°C; pH 7.0; 10.0 min reaction time	11.16	16.37	+5.21	Two‐step transglycosylation reconfigures starch into a dense branched network, establishing structural barriers that effectively restrict amylolytic access	Reduced molecular weight, preserved crystalline structure Diminished molecular weight and induced C‐type crystalline polymorphs	Xia et al. (2021)
Waxy corn	4,6‐α‐Glucanotransferase (GsGtfC)	Starch slurry 5%, w/v; 100 U/g substrate; 65°C; pH 5.5; 1.0 h reaction time	11.07	24.11	+13.04	GsGtfC transforms α‐1,4 bonds into α‐1,6‐branches, synthesizing reuteran‐like polymers that effectively restrict amylolytic access	Induced α‐1,6‐branching to create compact polymers; reduced molecular weight and crystallinity Minimized crystalline structures and molecular dimensions while maximizing chain density and short‐branch fractions	Li, Wang, et al. (2022)

Note: Comparisons across studies should be interpreted cautiously because digestion protocols, sample preparation, and starch fractionation methods may differ.

Abbreviation: DP, degree of polymerization.

^a SDS values are reported as presented in the cited studies.

^b Absolute change in SDS was calculated as SDS after–SDS before (%).

4.7. Limitations of the Current Evidence: Window Boundaries and Unreported Stability

The pattern described across Sections 4.1–4.6, that SDS gains favor an intermediate degree of enzymatic modification rather than maximal enzyme activity, is a synthesis drawn from studies whose primary aim was usually something other than SDS optimization. Most of the cited work targeted RS formation or a general reduction in digestibility, with SDS reported as one outcome among several rather than the variable being deliberately maximized. Across most enzyme–starch combinations, neither the full range of conditions nor the range of botanical sources has been tested, so the location of the reversal point remains inferred rather than mapped. A further limitation applies across all studies reviewed in this section: There is a lack of knowledge on whether the SDS formed persisted through storage, reheating, or subsequent processing. Combined physical–enzymatic strategies, discussed in Sections 5 and 6, may offer a possible route to further enhancing SDS formation or stability if conditions are optimized.

5. Combined Physical–Enzymatic Modification Strategies

Sequential integration of enzymatic and physical treatments has emerged as a particularly effective approach for modifying starch digestibility because it couples molecular restructuring with supramolecular reorganization (Li, Gui, et al. 2020). Specifically, physical methods, such as extrusion, HMT, or autoclaving, can serve as essential pretreatments by disrupting the dense, semicrystalline structure of native starch granules. This disruption improves enzyme accessibility, as native granules often restrict enzyme penetration, limiting the efficiency of debranching (Doan et al. 2025; Song et al. 2026). Depending on the modification strategy, physical pretreatment may enhance subsequent enzymatic reactions, or enzymatic debranching may first generate linear glucan chains that are subsequently reorganized through physical processing. Together, these complementary treatments modify both the molecular architecture and supramolecular organization of starch, promoting the formation of stable ordered structures with reduced enzymatic accessibility (Li, Gui, et al. 2020).

Reported ΔSDS values vary considerably, ranging from moderate increases to improvements exceeding 40 percentage points, depending on starch botanical origin, treatment sequence, and processing conditions (Table 4). Across different starch sources, combined enzymatic–physical treatments consistently produce greater SDS increases than either treatment alone, although the magnitude of improvement differs among systems. For instance, combining extrusion with pullulanase debranching significantly increases the SDS content of corn starch from 7.04% to 25.89%; a similar trend is observed in potato starch, where the SDS fraction increases from 7.81% to 30.11% (Liu et al. 2022). In rice starch, the SDS content rises from 9.12% to 25.68% when extrusion is followed by pullulanase treatment (Song et al. 2026). Compared with physical treatments alone, enzymatic restructuring directly remodels starch molecular architecture, whereas subsequent physical processing promotes the reorganization of these modified chains into more ordered supramolecular structures that enhance SDS formation (Fan et al. 2023).

TABLE 4.

Dual modification strategies to enhance slowly digestible starch (SDS).

Starch source	Enzyme treatment	Experimental conditions	SDS before (%) a	SDS after (%) a	Absolute change in SDS (%) b	Experimental mechanism	Reference
Sweet potato	Pullulanase + dual retrogradation	10% (w/v) starch; 20 U/g enzyme; 55°C; pH 5.0; 5.0 h (hydrolysis) + dual‐retrogradation (4°C; 48.0 h per cycle)	15.41	57.82	+42.41	Pullulanase debranching releases linear glucans that reorganize into imperfect crystallites via dual‐retrogradation, enhancing digestion resistance	Luong et al. (2024)
Waxy rice	Pullulanase + recrystallization	10% w/v starch; 60 ASPU/g pullulanase; pH 5.2; 58°C; 12 h (debranching); 80% moisture; 4 cycles of boiling (30 min) followed by recrystallization (48 h at 4°C)	45.50	57.80	+12.30	Pullulanase debranching and four‐cycle crystallization reorganize short linear chains into stable double helices and imperfect crystallites	F. Zeng, Chen, et al. (2015)
Sweet potato	Pullulanase + heat–moisture treatment	11.11% w/v starch (5.0 g/45 mL); 25 ASPU/g enzyme; pH 4.4; 58°C for 24 h (debranching); followed by HMT at 100°C for 2 h with 30% moisture content	7.71	31.60	+23.89	Reorganize short linear chains into perfect double helices, increasing crystallite stability	Huang et al. (2015)
Rice	Autoclaving + pullulanase	20% (w/v) starch; autoclaved (121°C; 1 h); 20 U/g pullulanase; 60°C; pH 5.0; 6.0 h reaction; 24.0 h retrogradation at 4°C	7.20	23.20	+16.00	Autoclaving releases amylose while pullulanase debranches amylopectin to enhance molecular mobility, facilitating the alignment and aggregation of linear chains into ordered B‐type double helices	Li, Gui, et al. (2020)
Yellow Pea	Pullulanase + ultrasonication	5% (w/v) starch; precooked (80°C; 10 min); 40 npun/g pullulanase; simultaneous ultrasonication (100% amp; 1 min on/9 min off); 50°C; pH 5.2; 24.0 h duration	5.50	17.10	+11.60	Synergistic pullulanase debranching and ultrasonic scission liberate linear glucans that reorganize into digestion‐resistant retrograded networks	Lu et al. (2018)
Potato	Extrusion + pullulanase	10% w/v extruded starch; 40 PUN/g starch enzyme; pH 4.6; 58°C; 8 h; followed by recrystallization (4°C, 24 h)	7.04	25.89	+18.85	Extrusion‐assisted debranching produces short linear chains that rearrange into stable double helices, increasing thermal stability	Liu et al. (2022)
Banana	Heat–moisture treatment + α amylase and glucoamylase	HMT 30% moisture (20 h equilibration); two cycles of 100°C for 5 h each; 10% w/v starch suspension (pH 6.0); 8 U/g α‐amylase and 24 U/g glucoamylase; 50°C for 9 h; dried at 50°C	21.72	26.95	+5.23	Sequential modification facilitates molecular rearrangement and concentrates crystalline fractions by enhancing enzyme accessibility to the amorphous domains for amylose degradation	Garofalo et al. (2024)
Potato	α‐Amylase and pullulanase	6.25% w/v starch; 5 U/g α‐amylase; pH 5.0; 55°C; 6 h; followed by 40 U/g pullulanase; pH 5.3; 60°C; 10 h; retrograded at 4°C for 16 h	18.00	27.80	+9.8	Reduced the chain length of long molecular fractions, which upon debranching, assemble into imperfect B‐type crystallites	Villas‐Boas and Franco (2016)
Corn	Extrusion + recrystallization	60% extrusion moisture; 40–120°C temp.; 7‐day recrystallization at 4°C; dried at 40°C	7.04	25.62	+18.58	Extrusion‐induced thermomechanical shear selectively degrades amylopectin into flexible, low‐molecular‐weight chains that readily rearrange into stable crystallites during storage	Liu et al. (2021)

Note: Comparisons across studies should be interpreted cautiously because digestion protocols, sample preparation, and starch fractionation methods may differ.

Abbreviation: HMT, heat–moisture treatment.

^a SDS values are reported as presented in the cited studies.

^b Absolute change in SDS was calculated as SDS after–SDS before (%).

Mechanistically, the effectiveness of these hybrid treatments is largely associated with the generation of linear or moderately branched glucan chains during enzymatic debranching reactions. The simultaneous use of pullulanase and ultrasonication provides a clear example of this synergy. Ultrasound creates a mechanochemical effect that lowers system viscosity and opens the branched amylopectin structure, thereby improving enzyme accessibility and enhancing pullulanase action (Song et al. 2026). In pea starch, pullulanase alone increased SDS from the precooked substrate's baseline to 5% after 6 h, while adding simultaneous ultrasonication under the same conditions raised SDS to 18% over the same 6‐h period (Lu et al. 2018). The linear glucan chains generated by pullulanase subsequently exhibit an enhanced ability to reassociate into double‐helical structures during cooling or retrogradation. Physical processing steps then facilitate the alignment and packing of these chains into more stable crystalline domains, often producing compact structures with reduced enzyme penetration. For example, rice starch treated with autoclaving and a sequential triple‐enzyme process develops a dense, solid structure where surface pores almost entirely disappear. This specific modification increases the SDS level to 18.6% and reduces the glycemic index to 62.9 (Li, Gui, et al. 2020). However, increasing treatment complexity does not necessarily maximize SDS formation. Although some dual‐modification strategies substantially enhance SDS, more intensive enzymatic treatments may preferentially promote RS formation, indicating that an optimal balance between chain remodeling and structural reorganization is required. Such treatments can transform moderately digestible starch systems into matrices containing substantially higher SDS fractions (Babu et al. 2019; Yang et al. 2025).

The effectiveness of combined modification strategies therefore reflects the interaction between two complementary processes: control of molecular architecture and physical promotion of supramolecular ordering. Enzymatic treatments generate glucan chain architectures favorable for molecular reassociation, whereas subsequent physical processing stabilizes these chains into partially ordered structures responsible for slow digestion. Consequently, the final digestibility profile depends not only on the individual treatments employed but also on their sequence, intensity, and the botanical origin of the starch. By simultaneously influencing both levels of starch structure, these approaches provide a powerful means of tailoring starch digestibility and generating more stable slowly digestible fractions (Hutabarat and Stevensen 2023; Li et al. 2019).

6. Comparative Effectiveness of Modification Strategies for Enhancing SDS

The studies summarized in Tables 1, 2, 3, 4 reveal that the effectiveness of different modification strategies in enhancing SDS depends primarily on the extent and scale of structural transformation induced within the starch matrix. Across the reviewed treatments, increases in SDS are associated with modifications occurring at multiple structural levels, including molecular architecture, crystalline organization, and granule morphology. The relative contribution of these structural changes varies among physical, enzymatic, and combined modification approaches.

Physical modification strategies primarily influence the supramolecular organization of starch granules. Treatments that induce structural disruption sufficient to enable reassociation into partially ordered, enzyme‐resistant domains tend to generate the largest increases in SDS; however, this relationship is non‐monotonic, disruption exceeding a starch‐ and method‐specific threshold redirects the outcome toward RDS or RS rather than producing further SDS gains. Mechanical treatments, such as ball milling, illustrate this bidirectionality directly: Controlled milling of high‐RS starches, such as potato, can convert resistant structures into SDS (Lv et al. 2019), whereas equivalent or more intensive milling of lower RS starches, such as wheat, instead depolymerizes SDS toward RDS (Han et al. 2022), whereas hydrothermal treatments, such as HMT or microwave processing, primarily operate through gradual rearrangement of crystalline and amorphous regions. Consequently, the magnitude of SDS enhancement observed for physical treatments is strongly dependent on processing intensity, post‐process storage condition, and the structural characteristics of the starch source (Gou et al. 2019; Wang et al. 2019).

In contrast, enzymatic modification alters starch digestibility through direct restructuring of molecular architecture. By selectively cleaving or redistributing glycosidic linkages, enzymatic treatments modify the chain‐length distribution and branching pattern of amylopectin, which strongly influences the formation and stability of double‐helical structures. These molecular changes can significantly affect starch digestion behavior even in cases where granule morphology remains largely intact. The resulting SDS levels therefore depend primarily on the type of enzymatic reaction and the degree to which the resulting glucan chains can reorganize into ordered structures during cooling or storage (Jiang et al. 2014; H. Li, Gui, et al. 2020; Li, Fei, et al. 2020).

Combined modification strategies integrate these two mechanisms by coupling enzymatic control of molecular architecture with physical promotion of supramolecular ordering. Such sequential treatments enable enzymatic reactions to generate chain configurations favorable for crystallization, whereas subsequent physical processing facilitates the alignment and packing of these chains into stable crystalline domains (Li, Fei, et al. 2020). As reflected in Table 4, this interaction between molecular restructuring and granular reorganization can produce particularly pronounced increases in SDS.

The comparative evidence indicates that the magnitude of SDS enhancement is closely linked to the degree of hierarchical structural reorganization achieved within the starch matrix. Modification strategies that influence both molecular architecture and supramolecular organization tend to produce the most substantial changes in digestibility, whereas treatments affecting only a single structural level generally result in more moderate effects. These findings highlight the importance of considering starch structure across multiple scales when designing processing strategies aimed at controlling starch digestion behavior.

7. Structural Determinants Governing SDS Formation

Despite the diversity of processing strategies used to modify starch digestibility, the studies reviewed consistently indicate that increases in SDS arise from a limited number of structural transformations within the starch matrix, operating across molecular, crystalline, and granule‐level scales (Su et al. 2020).

The central determinant is the stage reached in the starch crystallization pathway. SDS and RS arise from the same underlying process of chain reassociation but are arrested at different points along it. Processing conditions that favor rapid nucleation, promoting the formation of numerous small, imperfect crystallites, generate SDS. Conditions that instead allow those nuclei to propagate and mature into larger, more perfect crystalline domains push the same process toward RS. This distinction places SDS not as a lesser degree of crystalline order than RS, but as crystallization deliberately arrested at the nucleation stage, before propagation and maturation occur (Li, Zhang, et al. 2022; Li et al. 2019). Where processing severity instead disrupts the starch matrix beyond the point at which chains can reassociate, seen repeatedly across DHT, ball milling, cold plasma, and high‐dose enzymatic treatment earlier in this review, the outcome bypasses both SDS and RS entirely, yielding RDS instead. SDS formation therefore depends on reaching a structural state where reassociation is possible before determining how far along the nucleation‐to‐maturation axis that reassociation proceeds.

Chain length exerts a further, largely independent effect. Short chains and noncrystalline amylose are digested rapidly, whereas intermediate and longer chains impede hydrolysis through an anchoring or entanglement effect on neighboring chain segments, distinct from any contribution of crystalline packing. This chain‐length effect can therefore modulate digestibility even where crystallinity itself is held constant (Miao et al. 2015).

Branching density adds a third, non‐unidirectional layer. Both increasing branching (via branching enzyme treatment) and decreasing it (via debranching) have been shown to raise SDS content, indicating that branching density does not act on digestibility in a single direction. Its effect instead depends on whether the resulting change in branch architecture is accompanied by crystalline reorganization, debranching promotes SDS primarily by liberating short chains for realignment into crystallites, whereas added branching can restrict enzyme access directly through increased α‐1,6 linkage density, independent of crystallinity change (Miao et al. 2015; Su et al. 2020).

Granule and matrix architecture contribute a structurally separate determinant. Treatments that restrict granule swelling or generate dense, compact structural domains reduce the physical space available for enzyme entry, independently of the degree of crystalline order present (Liu et al. 2019; Su et al. 2020).

This review is restricted to structural transformations within native starch subjected to physical or enzymatic modification alone; interactions between starch and other food matrix components fall outside this scope and are not addressed here. However, the significant effects of such interactions on starch digestibility, particularly in real food systems, should not be overlooked when translating these structural principles to practical food applications.

SDS formation is therefore governed by four distinguishable determinants: The stage reached in the crystallization pathway (nucleation vs. maturation), chain length, branching density, and granule or matrix architecture. Modification strategies capable of influencing several of these determinants simultaneously tend to produce the most pronounced effects on starch digestibility. Understanding these determinants provides an important framework for designing processing strategies aimed at controlling starch digestion behavior and developing foods with improved glycemic properties (Magallanes‐Cruz et al. 2017; Shi et al. 2018).

8. Emerging Perspectives in Enzymatic Engineering of SDS

Beyond conventional enzymatic hydrolysis approaches, recent advances highlight that enzymatic modification can be strategically employed to engineer starch structure at multiple hierarchical levels, thereby modulating digestibility in a more targeted manner. Rather than solely reducing molecular size, specific enzymes can be used to reconfigure chain architecture, branching patterns, and supramolecular organization, which are critical determinants of SDS formation.

Building on the debranching mechanisms discussed in Section 4, recent studies have shifted from simply increasing debranching efficiency toward precisely controlling chain‐length distribution and molecular architecture to optimize SDS formation (Korompokis et al. 2021). Beyond debranching, transglycosylating enzymes such as amylomaltase, whose mechanism is discussed in Section 4.5, remain an active area for enzyme discovery and reaction‐condition optimization as this reorganization‐driven strategy matures toward practical application.

Another underexplored aspect is the role of enzymatic modification in influencing enzyme–substrate interactions during digestion itself. Structural features introduced through enzymatic treatments, such as altered branching density, reduced accessibility of α‐1,4 linkages, or increased formation of junction zones, can significantly affect the binding and catalytic efficiency of digestive enzymes. For example, α‐amylase preferentially attacks amorphous regions and requires sufficient chain alignment for effective binding, whereas mucosal α‐glucosidases can further hydrolyze residual structures, including those not readily accessible to α‐amylase. Designing starch structures that limit these interactions provides an additional pathway to regulate digestion kinetics (Korompokis et al. 2021).

Furthermore, enzymatic treatments can promote the formation of intermediate structures that evolve during digestion, such as linear glucan fragments that reassociate into double helices in situ, contributing to reduced digestibility. This dynamic restructuring during digestion suggests that enzymatic modification should not only be considered in terms of preprocessing effects but also in relation to structural transformations occurring under gastrointestinal conditions.

Advanced enzymatic strategies enable precise tuning of molecular architecture and supramolecular organization, shifting the focus from simple hydrolysis to the rational design of digestion behavior. Integrating enzyme specificity, reaction conditions, and posttreatment structural evolution provides a robust framework for developing starch systems with enhanced SDS content and controlled glycemic response.

9. Industrial Translation of SDS Knowledge: Commercial Ingredients and Applications

Although SDS has been widely investigated in academic studies, its translation into commercially available ingredients remains relatively limited. As shown in Table 5, only a small number of commercial products are specifically developed or marketed as SDS ingredients, with SUSTRA 2434 (51% SDS, Englyst Assay) and Pea Starch LN30 (30% SDS, combined with 54% RS) representing the clearest examples. The remaining commercially available starch‐based functional ingredients are positioned as RS, resistant maltodextrin, or dietary fiber systems rather than SDS, highlighting a substantial gap between advances in SDS research and their industrial implementation.

TABLE 5.

Examples of commercially available slowly digestible starch (SDS) and resistant starch (RS) ingredients used to modulate postprandial glycemic response in food and nutraceutical products.

Company	Country	Starch source/product	Digestibility classification	Functional application	Reference
Ingredion Incorporated	USA	SUSTRA 2434, tapioca flour/corn starch blend	SDS (51% SDS, Englyst Assay)	Slowly digestible carbohydrate; sustained post‐meal energy	Gourineni et al. (2017)
Roquette Frères	France	Pea Starch LN30, native pea starch	SDS + RS (30% SDS, 54% RS)	Native starch; lowers glycemic response	Perreau et al. (2023)
Ingredion Incorporated	USA	Hi‐Maize 260, high‐amylose maize starch	RS2	Dietary fiber; supports glycemic and renal health	Headley et al. (2025)
Ingredion Incorporated	USA	VERSAFIBE 1490, potato‐based distarch phosphate	RS4	Dietary fiber; reduces postprandial glucose and insulin	Stewart and Zimmer (2017)
Cargill, Inc.	USA	Actistar 75330, chemically modified tapioca starch	RS4	Dietary fiber substitution; lowers postprandial glucose and insulin	Mah et al. (2018)
MSP Starch Products Inc.	Canada	Solnul, unmodified resistant potato starch	RS2	Nutraceutical; prebiotic effect	Bush et al. (2023)
ADM/Matsutani LLC	USA/Japan (joint venture)	Fibersol‐2, resistant maltodextrin from cornstarch	Resistant maltodextrin (soluble dietary fiber)	Dietary fiber; reduces visceral fat and supports metabolic health with continuous consumption	Namkieat et al. (2025)

Commercial strategies for modulating starch digestibility fall into two broad categories. The first comprises ingredients designed for sustained glucose release, used in sports nutrition, medical nutrition, and functional beverages where prolonged energy delivery is desirable. The second, and by far the more established category, comprises RS and high‐amylose starch ingredients used to reduce digestibility and glycemic response; high‐amylose maize starch in particular is widely incorporated into bread, pasta, cereal, and snack products to increase fiber content and lower glycemic impact without major formulation changes or loss of sensory quality.

Although these RS‐based products can contribute to slower glucose release or reduced postprandial glycemic responses, they do not specifically target the production or maintenance of high SDS fractions, underscoring that current industrial innovation remains focused on increasing RS rather than deliberately engineering stable SDS structures.

A further challenge is that SDS content is often influenced by subsequent food processing rather than being an inherent property of the ingredient itself. Thermal processing, extrusion, retrogradation, cooling, and interactions with other food components can substantially alter starch structure, meaning that the SDS content measured in an isolated ingredient may not be retained in the final food product. Producing ingredients capable of maintaining stable SDS levels throughout industrial processing therefore remains a major technological challenge.

This gap represents an important opportunity for future ingredient development. Manufacturers increasingly seek carbohydrate systems capable of delivering sustained energy while maintaining clean labels, desirable sensory properties, and processing stability, an advantage SDS offers over RS by providing sustained release without the extremely low digestibility associated with RS. Realizing this potential will require improved understanding of starch molecular architecture, ingredient interactions, and processing‐induced structural transformations, bridging the current gap between fundamental starch science and the commercial production of next‐generation SDS ingredients.

10. Conclusion and Future Perspectives

The present review demonstrates that physical, enzymatic, and their combined modification strategies provide effective green, safe, and nonchemical approaches for enhancing the proportion of SDS in starch‐based systems. Across the reviewed studies, increases in SDS consistently arise from structural transformations occurring at multiple hierarchical levels, including molecular architecture, crystalline organization, and granule or matrix structure.

Physical treatments primarily influence supramolecular organization through disruption and subsequent reorganization of starch granules, whereas enzymatic restructuring directly modifies glucan chain architecture and promotes the formation of ordered double‐helical structures. The sequential integration of these approaches produces synergistic effects by coupling molecular restructuring with enhanced supramolecular ordering. Accordingly, combined physical–enzymatic treatments generally yield the greatest increases in SDS compared with each individual method, with ball milling of potato starch (Table 1) as a notable exception, its exceptionally high single‐treatment gain reflects a distinct fragmentation‐driven mechanism in RS‐rich starches rather than the gradual disruption‐reorganization pathway that governs most single and combined treatments alike. Single physical or enzymatic modifications generally result in more moderate improvements depending on processing conditions and starch type.

However, an increase in SDS does not always reflect a favorable digestibility outcome: Several modification strategies raise SDS alongside a concurrent increase in RDS, underscoring the need for careful, outcome‐specific assessment of each method rather than SDS content in isolation. The most desirable modification strategies are those that simultaneously increase SDS, reduce RDS, and increase RS.

Future research should move beyond intrinsic starch structural modifications to consider the role of multicomponent food matrices in governing SDS formation. In real food systems, interactions between starch and other components, including hydrocolloids, proteins, lipids, and polyphenols, can significantly influence digestion behavior by modifying matrix structure, stabilizing ordered domains, and restricting enzyme diffusion. These matrix‐level effects highlight the need to integrate starch modification strategies with formulation‐based approaches that reflect realistic food systems.

Accordingly, future studies should focus on SDS process tolerance and the development of combined processing–formulation strategies that account for ingredient interactions at both particle and molecular levels. Advances in structural characterization techniques, such as synchrotron x‐ray scattering, solid‐state NMR, and advanced microscopy, will be critical for elucidating the multi‐scale mechanisms underpinning SDS formation and enabling their translation into practical, industrially relevant food applications.

Developing integrated processing and formulation strategies capable of controlling starch digestion behavior will be essential for producing next‐generation carbohydrate‐rich foods with improved metabolic functionality. Such approaches offer significant opportunities to design starch‐based products that deliver sustained glucose release while maintaining desirable technological and sensory properties.

Author Contributions

Samiah Khalaf Alshammari: writing – original draft, methodology, writing – review and editing. Asgar Farahnaky: conceptualization, methodology, writing – review and editing, supervision. Ayman Allahham: methodology, writing – review and editing, supervision. Billy Lo: writing – review and editing, methodology, validation. Nilesh Nirmal: writing – review and editing, methodology, validation. Mahsa Majzoobi: conceptualization, writing – review and editing, supervision, methodology, validation, data curation.

Funding

The PhD Scholarship was funded through project number “NBU‐SAFIR‐2026”.

Conflicts of Interest

The authors declare no conflicts of interest.