Skip to main content
NCBI home page
RSC Advances logoLink to RSC Advances
. 2026 Jul 9;16(36):38870–38879. doi: 10.1039/d6ra00162a

Perioperative serum ALP antigen profiling to compare unilateral and bilateral vertebroplasty in osteoporotic vertebral compression fractures using an electrochemical immunosensor

Yinfei Xu b, Ying Chen a, Maoyuan Bian a, Cai Wang a,✉, Jiuming Dai a,✉
PMCID: PMC13347291  PMID: 42427412

Abstract

This study aims to establish an electrochemical immunosensor based on silver nanocubic particles (AgNPs) for the dynamic monitoring of serum alkaline phosphatase (ALP) antigen levels in patients with osteoporotic vertebral compression fractures (OVCF) during the perioperative period. This platform allowed a preliminary exploration of postoperative ALP antigen trajectories in patients receiving unilateral or bilateral percutaneous vertebroplasty (PVP). Experimental results demonstrate that this sensor exhibits excellent linear response, low detection limits (1.53 pg mL−1 and 1.74 pg mL−1 respectively), outstanding stability, repeatability, specificity, interference resistance, and clinical applicability. Exploratory clinical sample analysis showed that serum ALP antigen levels in the bilateral injection group tended to be higher than those in the unilateral injection group at one month post-surgery, suggesting that different injection strategies may be associated with distinct postoperative ALP antigen trajectories related to bone metabolism. However, this observation should be interpreted cautiously and requires further validation in larger cohorts with appropriate statistical testing and adjustment for potential confounders. This sensor demonstrates potential for application in biosensing and clinical diagnostic research.

This study establishes an silver nanocubic particles based electrochemical immunosensor for dynamic monitoring of serum alkaline phosphatase antigen in osteoporotic vertebral compression fractures patients perioperatively.graphic file with name d6ra00162a-ga.jpg

1. Introduction

Thoracolumbar compression fractures, particularly osteoporotic vertebral compression fractures (OVCF), may result in significant adverse effects for patients, including pronounced pain, restricted mobility, and diminished quality of life. Driven by clinical demands for pain management and vertebral stabilisation, percutaneous vertebroplasty (PVP) has emerged as a significant minimally invasive intervention.1 However, the impact of PVP on genuine clinical benefits, suitable patient populations, and long-term outcomes remains subject to ongoing debate. A randomised controlled trial study revealed that the efficacy of PVP is not consistent across all scenarios, being closely associated with factors such as the enrolment time window, radiographic bone marrow oedema, and duration of pain.2,3 Moreover, within the surgical technique pathways for PVP, unilateral puncture and bilateral puncture represent the two most common surgical approaches. Previous studies have demonstrated significant differences between the two surgical approaches in terms of operative duration, fluoroscopic exposure, bone cement consumption, and distribution patterns. They also frequently exhibit distinct characteristics regarding pain management, functional outcomes, and complications such as bone cement leakage.4 This suggests that relying solely on imaging and quantitative outcome measures may be insufficient to address the more clinically and biologically pertinent question of whether different puncture strategies elicit differentiated bone repair responses. Therefore, in addition to traditional assessment metrics such as pain scales, functional scores, and radiographic stability, there is an urgent need to introduce more refined, reproducible perioperative evaluation indicators capable of reflecting the ‘biological process of bone repair’. This will address the limitations of relying solely on clinical outcomes, which inadequately explain differences at the mechanistic level.

Alkaline phosphatase (ALP) plays a pivotal role in bone mineralisation. By regulating the microenvironmental balance of inorganic phosphate and pyrophosphate, it participates in mineral deposition and hard tissue formation, making it one of the most commonly employed biochemical markers in osteogenesis-related research.5,6 It should be emphasised that bone repair activity following fracture exhibits distinct temporal characteristics, with the molecular network undergoing dynamic fluctuations throughout the inflammatory-regenerative-remodelling phases.5 From the perspective of surgical technique variation research, if a particular strategy facilitates bone formation-related responses while achieving mechanical stability, the corresponding bone metabolism biomarker trajectory may exhibit detectable, reproducible differences. Therefore, monitoring changes in patients' ALP levels following PVP administration at OVCF demonstrates clear biological rationale and clinical translational potential.6 It should be noted that the present immunosensor quantifies ALP antigen concentration based on immunorecognition rather than ALP enzymatic activity. Therefore, ALP-related biological functions are discussed here as the biological background of this biomarker, while the analytical readout in this study should be interpreted as the serum ALP antigen level.

The predominant method for measuring ALP in clinical laboratories remains the kinetic enzymatic assay utilising automated biochemical analysis platforms. Typically, based on the p-nitrophenyl phosphate substrate system, the reaction rate is measured under strictly controlled temperature, buffer conditions, and wavelength parameters to characterise the catalytic activity concentration of ALP.7 However, ALP activity results may be influenced by isoenzymes from different tissue sources. In monitoring scenarios involving low sample volumes and rapid feedback, time constraints, accessibility, and instrument dependency remain practical limitations.7,8 Therefore, the development of novel testing technologies that are miniaturisable, capable of rapid quantitative analysis, and compatible with clinical blood collection procedures holds significant clinical importance.

In recent years, novel detection technologies for point-of-care testing (POCT) have continued to emerge. Although methods such as paper-based platforms, surface-enhanced Raman spectroscopy (SERS), and electrochemiluminescence (ECL) have expanded the range of ALP detection approaches, their limitations remain pronounced when applied to perioperative blood biomarkers requiring multi-time-point, repeatable, and rapid feedback.8–10 Paper-based platforms are susceptible to variations in paper batches and environmental factors, resulting in limited quantitative accuracy and batch consistency. They frequently require external readers and standardised calibration procedures.11,12 Although SERS possesses ultra-sensitive potential, signal intensity remains highly dependent on the consistency between the “hot spot” and the substrate. Achieving quantitative reproducibility across batches and laboratories remains a core technical bottleneck, necessitating higher demands on both instrument platforms and sample pretreatment protocols.13,14 ECL combine high sensitivity with a broad dynamic range and are well-established in industrial applications. However, they rely on optical detection and relatively complex luminescent labelling strategies, whilst their equipment costs and operational barriers cannot be overlooked.15,16 By contrast, ALP antigen concentration quantification based on immunological identification is more suitable for clinical protein biomarker testing. Electrochemical immunosensing is emerging as a key technology for rapid quantitative biomarker detection due to its direct signal conversion, ease of miniaturisation and integration, and compatibility with microelectrode systems.17–19 Silver-based nanomaterials, owing to their excellent conductivity and potential for interfacial chemical modification, have been systematically summarised as a crucial functional material system for immunoelectrochemical detection.20

This study established an electrochemical immunosensing platform for quantifying ALP antigen concentrations in blood samples (Scheme 1). The platinum–carbon electrode (PC) provides a stable conductive substrate and ensures electrode consistency. Silver nanocube particles (AgNPs), owing to their well-defined nanostructure and excellent conductivity, provide abundant interfacial immobilization sites and facilitate charge-transfer processes at the electrode surface. 5,5′-Dithiobis(2-nitrobenzoic acid) (DTNB) serves as a multifunctional interfacial molecule, contributing to AgNPs surface immobilisation, EDC/NHS-mediated antibody coupling through its carboxyl groups, and DTNB-derived electrochemical signal generation. Owing to its sulfur-containing/disulfide structure, DTNB can be introduced onto the AgNP-modified surface through the strong affinity between sulfur-containing moieties and silver surfaces, which is generally associated with Ag–S interfacial interactions. This facilitates the construction of electrochemical readout layers on electrode surfaces. To covalently couple anti-ALP antibodies to the interface for specific immunorecognition, (1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride/N-hydroxysuccinimide (EDC/NHS)) was employed to convert surface carboxyl groups into intermediates capable of reacting efficiently with protein amines. Finally, non-specific adsorption and matrix interference are reduced by blocking unbound sites with bovine serum albumin (BSA). This design converts antigen–antibody binding events into measurable differential pulse voltammetry (DPV) signal variations, enabling quantitative detection of ALP antigen concentrations within clinically relevant ranges. This study employed the sensor to monitor ALP antigen concentrations at various post-operative time points following PVP procedures in OVCF patients, as well as across different surgical approaches. This study employed the sensor to monitor perioperative serum ALP antigen levels in OVCF patients undergoing PVP and to preliminarily explore whether different injection strategies were associated with distinct postoperative ALP antigen trajectories. These findings provide preliminary analytical evidence supporting the feasibility of ALP antigen monitoring in perioperative clinical samples.

Scheme 1. Schematic diagram of the electrochemical immunosensor fabrication and detection process.

Scheme 1

Open in a new tab

2. Materials and methods

2.1. Materials

5,5′-Dithiobis(2-nitrobenzoic acid) (DTNB), 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride/N-hydroxysuccinimide (EDC/NHS), polyvinylpyrrolidone, hexadecyltrimethylammonium chloride (CTAC), silver nitrate (AgNO3), sodium borohydride purchased from Aladdin Industrial Co., Ltd, Shanghai, China. Cetyl Trimethyl Ammonium Chloride (CTAC), sodium borohydride were procured from Aladdin Industrial Co., Ltd, Shanghai, China. Hydrochloric acid (HCl), potassium ferricyanide (K3[Fe(CN)6]), potassium chloride (KCl), ascorbic acid, bovine serum albumin (BSA), glutathione (GSH), lysine (Lys), and glycine (Gly) were all procured from the National Pharmaceutical Group Chemical Reagents Co., Ltd of the United States. Interleukin-2 antigen (IL-2), IL-4 antigen, neuron-specific enolase (NSE), glial fibrillary acidic protein (GFAP), carcinoembryonic antigen (CEA), alkaline phosphatase (ALP), and antibodies were all purchased from Sigma-Aldrich Trading Co., Ltd, Shanghai, China. All purity grades exceed 99.9%.

2.2. Instruments

The electrochemical workstation (CHI660E), platinum–carbon electrode, reference electrode and counter electrode were all procured from Shanghai Chenhua Instruments Co., Ltd. The PB21 acidity meter was purchased from Sedo-Rius Scientific Instruments (Beijing) Co., Ltd. The scanning electron microscope employed for morphological characterisation of the nanoparticles was a FE-SEM, GeminiSEM 300 (Carl Zeiss, Germany). Ultraviolet-visible (UV-Vis) absorption spectra were measured using a UV-Vis spectrophotometer (SHIMADZU UV-3600plus, Japan). Ultrapure water for experimental use was filtered through a Milli-Q reagent water system (Millipore, USA).

2.3. Preparation of AgNPs

First, at 30 °C, a freshly prepared 0.01 M sodium borohydride aqueous solution was rapidly added to a mixed aqueous solution containing 0.10 M cetyltrimethylammonium chloride (CTAC) and 0.10 M ascorbic acid under gentle stirring. Subsequently, an aqueous solution containing 0.01 M silver nitrate and 0.10 M CTAC was added, and the mixture was allowed to react for 30 min to obtain an Ag seed solution. Then, a measured volume of the Ag seed solution was added to a freshly prepared growth medium containing 0.10 M CTAC, 0.10 M ascorbic acid, and 0.01 M silver nitrate. After gentle mixing, the reaction was maintained at 30–35 °C for 4–10 h to allow the seed-mediated growth of AgNPs. After the reaction was completed, the resulting product was purified by centrifugation at 8000 rpm for 10 min and resuspended in deionised water. This washing step was repeated once or twice to obtain a monodisperse colloidal dispersion of AgNPs. Throughout this process, CTAC acted as a morphology-directing agent, with CTA+ selectively adsorbing onto silver crystal planes to regulate growth orientation, while Cl− modulated silver ion reduction and deposition kinetics. Ascorbic acid served as a mild reducing agent to enable the controlled deposition of silver atoms, thereby facilitating the formation of cubic AgNPs with sharp edges.21

2.4. Fabrication of the immunosensor

The pretreated platinum–carbon electrode was placed in a clean environment, where it was first modified with silver nanocubic particles. An appropriate amount of AgNPs solution of known concentration was applied by drop coating onto the electrode surface and dried at room temperature. AgNPs were retained on the electrode mainly through physical adsorption and electrostatic interactions, forming a conductive nanostructured substrate layer for subsequent DTNB immobilisation and interfacial electron transfer. The electrode was immersed in a DTNB solution of a specific concentration and incubated in the dark. DTNB molecules were introduced onto the AgNP-modified electrode surface through the strong affinity between sulfur-containing/disulfide moieties and silver surfaces, which is generally associated with Ag–S interfacial interactions. In this sensing interface, DTNB served as a multifunctional molecule, contributing to AgNPs surface immobilisation, EDC/NHS-mediated antibody coupling through its carboxyl groups, and DTNB-derived electrochemical signal generation. After incubation, the electrode was gently rinsed with phosphate-buffered saline to remove physically adsorbed DTNB molecules and then dried under nitrogen. The voltammetric signal was mainly attributed to the surface-confined DTNB-derived electrochemical response, which could be modulated by subsequent antibody immobilisation and antigen recognition. The EDC/NHS adduct activated the terminal carboxyl group of the immobilised DTNB molecule, forming a reactive ester intermediate. Following a brief rinse with buffer solution, the ALP primary antibody (Ab) solution was immediately pipetted onto the electrode surface. Prolonged incubation at low temperatures facilitated an efficient amidation reaction between the amino groups on the antibody molecules and the activated carboxyl groups, thereby achieving the directed and stable immobilisation of the antibodies. BSA was thoroughly incubated onto the electrode surface to coat all uncoupled active sites, thereby minimising non-specific adsorption during subsequent detection procedures. After blocking, the electrode was thoroughly rinsed with buffer solution, completing the construction of the layer-by-layer modified electrode comprising PC/AgNPs/DTNB/ALP Ab/BSA.

3. Results and discussion

3.1. Characterisation of AgNPs

To verify and characterise the physicochemical properties of the prepared AgNPs, scanning electron microscopy (SEM) was employed to observe the morphology of the AgNPs. As shown in Fig. 1A, the AgNPs prepared in this experiment exhibit a cubic morphology with sharp edges and demonstrate a degree of monodispersity. Furthermore, the detection results of the ultraviolet-visible (UV-Vis) absorption spectrum indicated that the absorption peak of AgNPs occurs around 500 nm (Fig. 1B). As the dimensions of silver nanocubes increase or due to the intense electromagnetic field enhancement effect arising from their distinctive sharp edges, the surface plasmon resonance absorption peak undergoes a redshift, consistent with experimental expectations.22

Fig. 1. (A) Typical SEM images and (B) UV-Vis spectra of the prepared AgNPs.

Fig. 1

Open in a new tab

3.2. Characterisation of prepared immunosensors

To validate the effectiveness of the stepwise modification of the electrode surface and the changes in interfacial charge transfer kinetics, electrochemical impedance spectroscopy (EIS) and cyclic voltammetry (CV) were employed to characterise the electrodes at different modification stages. EIS is a sensitive technique employed to track the dynamics of interfacial electron transfer. The electron transfer resistance (Rct) within its Nyquist plot is directly correlated with the insulating properties of the electrode surface. A higher Rct value corresponds to a larger semicircular diameter, indicating greater resistance to electron transfer at the electrode/solution interface.23 The impedance curves of electrodes at different modification stages (a(PC), b(PC/AgNPs), c(PC/AgNPs/DTNB), d(PC/AgNPs/DTNB/ALP Ab)) in a 1.0 mM [Fe(CN)6]4−/3− solution were shown in Fig. 2A. The fitted Rct values of PC, PC/AgNPs, PC/AgNPs/DTNB, and PC/AgNPs/DTNB/ALP Ab were 65.95, 63.62, 72.51, and 72.64 Ω, respectively. Compared with the bare PC electrode, the Rct value decreased from 65.95 Ω to 63.62 Ω after AgNPs modification, indicating that AgNPs facilitated interfacial electron transfer. This may be attributed to the good conductivity of AgNPs and the increased effective electroactive surface area provided by the nanostructured interface. After DTNB immobilisation, the Rct increased to 72.51 Ω, suggesting the introduction of an organic molecular layer on the AgNPs-modified surface. A slight further increase in Rct was observed after ALP Ab conjugation, which was consistent with the formation of a relatively insulating protein recognition layer. These quantitative EIS results further supported the stepwise construction of the PC/AgNPs/DTNB/ALP Ab sensing interface. Following DTNB modification and antibody covalent conjugation, the Rct value increased relative to the AgNPs-modified electrode, indicating that the organic signal layer and biomolecular recognition layer were sequentially introduced onto the electrode surface. The systematic progressive transformation of the EIS spectrum aligns precisely with the anticipated construction process of the transition from a conductive to an insulating sensing interface. This provided robust electrochemical evidence for the successful completion of each modification step and the ultimate realisation of specific immune recognition.20 To perform quantitative analysis of electrochemical impedance spectroscopy data, this study employed a suitably modified Randles equivalent circuit model, utilising ZSimpWin software to conduct non-linear least-squares fitting of experimental data. The fitted values of Rct for electrodes at different modification stages were analysed and compared. The equivalent circuit comprised solution resistance, constant-phase element, electron transfer resistance, and Warburg impedance. The impedance is defined as ZCPE = 1/[Q(jω)n], where Q is the CPE constant and n is the dissipation factor (with values ranging from 0 ≤ n ≤ 1).

Fig. 2. (A) Nyquist plots of the EIS for each immobilisation step for a(PC), b(PC/AgNPs), c(PC/AgNPs/DTNB), and d(PC/AgNPs/DTNB/ALP Ab) in 1.0 mM [Fe(CN)6]4−/3− solution. The fitted Rct values of PC, PC/AgNPs, PC/AgNPs/DTNB, and PC/AgNPs/DTNB/ALP Ab were 65.95, 63.62, 72.51, and 72.64 Ω, respectively. (B) CVs of the modified electrodes at different scan rates: 10, 20, 50, 80, 100, 150, and 200 mV s−1. The inset in (B) showed the linear plots of oxidation peak current (Ipa) and reduction peak current (Ipc) versus scan rate, together with the corresponding linear fitting results and R2 values.

Fig. 2

Open in a new tab

To evaluate the kinetic characteristics of the sensor, the CV results of the sensor electrode at different scan rates were shown in Fig. 2B. With increasing scan rate from 10 to 200 mV s−1, the redox peak currents increased progressively. Although some peak-potential shifts were observed with increasing scan rate, the shift of Epa and Epc was not strictly monotonic for every CV trace, indicating that the modified interface exhibited a certain degree of non-ideal electrochemical behaviour. This slight deviation may be related to the surface-confined electroactive layer, capacitive background current, local interfacial polarisation, or minor iR drop at higher scan rates. Therefore, the discussion was revised to avoid overinterpreting the peak-potential shift. More importantly, both the oxidation peak current (Ipa) and reduction peak current (Ipc) showed good linear relationships with the scan rate in the range of 10–200 mV s−1, with R2 values of 0.98 and 0.99, respectively. This linear dependence of peak current on scan rate indicated that the electrochemical response of the modified electrode was mainly governed by a surface-controlled process. The decrease in electron-transfer resistance after AgNP modification indicates that the nanostructured silver layer facilitates interfacial electron transport. Moreover, the uniformly distributed cubic morphology observed by SEM provides abundant exposed surface sites for subsequent DTNB and antibody immobilization. Although the effective surface area was not directly quantified in this study, the combined SEM, EIS, and CV results support the beneficial role of AgNPs in constructing an electrochemically accessible sensing interface. It should be noted that the EIS and CV results mainly provide electrochemical evidence for the stepwise construction of the sensing interface rather than direct spectroscopic confirmation of Ag–S bond formation. The increase in Rct after DTNB incubation is consistent with the successful introduction of an organic electroactive layer onto the AgNP-modified electrode. Considering the well-documented affinity between sulfur-containing groups and silver surfaces, these results support the immobilisation of DTNB on the AgNP-modified interface. However, direct identification of Ag–S bonding by surface-sensitive techniques such as XPS or Raman spectroscopy was not included in the present study and will be valuable in future mechanistic investigations.

3.3. Optimisation of experimental conditions

The pH of the solution significantly influences the electrochemical behaviour of the immunosensor, as the activity of the immobilised protein may be affected by the acidity or alkalinity of the solution.24 In the optimisation experiments, ΔI was defined as the absolute change in peak current before and after ALP antigen incubation under the same testing condition. Specifically, ΔI = |I0 − IALP|, where I0 represents the peak current recorded in the blank solution without ALP antigen, and IALP represents the peak current recorded after incubation with a fixed concentration of ALP antigen. This definition was used for the optimisation of pH, temperature, antibody concentration, and incubation time. To optimise the reaction pH, the sensor was tested using square wave voltammetry (SWV) in a series of PBS buffers spanning pH 4.0 to 9.0. Experimental results indicate that the change in current (ΔI) increases with rising pH, reaching a peak at pH 6.5 before gradually diminishing (Fig. 3A). Therefore, pH 6.5 was selected as the optimised pH value for the detection solution to achieve high sensitivity. This experiment also investigated the effect of temperature on the sensor, as temperature is a critical factor affecting the activity of antibodies and antigens. As shown in Fig. 3B, with increasing temperature, ΔI undergo a change process where it first decreased and then increased. At 35 °C, ΔI reached its minimum value. When the temperature rised (exceeding 40 °C), ΔI gradually increased. This may be attributable to irreversible reactions occurring during the process due to elevated temperatures. Consequently, experiments indicate that 35 °C represents the optimal reaction temperature for the sensor's operation. Antibody concentration exerts a significant influence upon the sensor's interfacial performance and detection outcomes. Both excessively high and low antibody concentrations can affect antigen binding. Excessively high concentrations increase non-specific adsorption, thereby reducing effective binding rates, while excessively low concentrations result in insufficient binding sites.25 To investigate the optimal antibody concentration for sensor operation, sensors modified with antibodies at varying concentrations were detected via SWV (Fig. 3C). Experimental results indicated that as the concentration of ALP antibodies gradually increased, ΔI progressively decreased until it stabilises at a concentration of 30 mg mL−1. Beyond this concentration, ΔI exhibited no significant change, maintaining a stable trend. Therefore, an antibody concentration of 30 mg mL−1 was determined to be the optimal concentration. Furthermore, the antigen–antibody incubation time is also critical for sensor detection. As shown in Fig. 3D, the constructed sensor was incubated with an ALP solution of a specific concentration under the aforementioned optimised conditions for varying durations. Experimental results indicated that ΔI rapidly decreased and stabilised within 30 min, demonstrating that antigen–antibody binding had reached saturation. To ensure a complete reaction and enhance detection efficiency, all subsequent experiments employed an incubation period of 30 min.

Fig. 3. SWV responses expressed as ΔI under varying (A) pH levels, (B) temperatures, (C) ALP antibody (Ab) concentrations, and (D) incubation times. ΔI was calculated as |I0 − IALP|, where I0 and IALP represent the peak currents recorded before and after ALP antigen incubation, respectively.

Fig. 3

Open in a new tab

3.4. Characterisation of immunosensor performance

Stability and repeatability are key indicators for evaluating sensor detection performance. Superior stability and repeatability can effectively enhance efficiency and reduce costs. First, to verify the sensor's stability, the prepared electrochemical sensors underwent DPV testing after being stored for varying durations (Fig. 4A). The peak current values detected by the sensor at different times were compared against the reference value (the sensor's maximum value). Even after being stored for over a month, no significant deviation was observed. This indicated that the constructed sensor exhibited excellent stability. Furthermore, multiple tests were conducted on sensors prepared from same batches (Fig. 4B), with peak current values taken as absolute values and the relative standard deviation (RSD) calculated. Experimental results indicated no significant difference in peak current values, with an RSD of 8%, demonstrating that the prepared sensors exhibit good reproducibility.

Fig. 4. (A) DPV detection results for the prepared sensor after a period of storage. (B) Results of multiple DPV tests conducted on sensors prepared from same batches. (C) DPV detection results of the sensor in solutions containing different substances (ALP, CEA, IL-4, IL-2, BSA, GSH, Lys). (D) Test results following the dissolution of ALP and various mixed substances (AA, GSH, NSE, GFAP, IL-2) in buffer solution and serum respectively.

Fig. 4

Open in a new tab

Furthermore, to validate the sensor's specificity and its resistance to interference and feasibility in clinical sample applications, we conducted tests in both phosphate-buffered saline (PBS) and serum. As shown in Fig. 4C, the sensor detected substances including ALP, CEA, IL-4, IL-2, BSA, GSH, Lys, etc., each at the same concentration (dissolved in PBS buffer solution). Test results indicated that the sensor exhibited a distinct signal and a significantly higher current value only when detecting ALP solution. As shown in Fig. 4D, various combinations of substances dissolved in buffer and serum were also detected by the sensor. The concentration of ALP was identical in each solution, while the concentrations of all other substances were twice that of ALP. In both the buffer and serum solutions, the sensor produced distinct peak currents in solutions containing ALP. It was noteworthy that when ALP coexisted with high concentrations of interfering substances (ascorbic acid (AA), GSH, NSE, GFAP, IL-2), the peak current values showed no significant change compared to ALP alone. Moreover, although the sensor exhibited differing current values in buffers and serum (due to the complex constituents in serum), its specificity remained intact. The results above indicated that the prepared electrochemical sensor exhibited good specificity towards ALP. Certain common electroactive substances and proteins did not cause significant interference with the sensor. More notably, the sensor continued to demonstrate excellent specificity and robusted interference resistance even when simulated in serum that mimics the actual biological environment.26 This preliminary finding indicates the immense potential of the sensor in clinical translation applications.

3.5. Detection of ALP with the immunosensor

To evaluate the electrochemical analytical performance of the sensor, we conducted measurements on ALP under the aforementioned optimised conditions. First, we employed DPV detection to analyse different concentrations of ALP dissolved in PBS. Within the concentration range, the DPV oxidation peak current value of the electrochemical sensor increased with rising ALP concentration (Fig. 5A). The oxidation peak current value of the sensor exhibited a good linear relationship with ALP concentration (Y = −0.1096X + 1.7571, R2 = 0.9713). Subsequently, as shown in Fig. 5B, ALP dissolved in serum at various concentrations was also detected by the sensor. The trend between concentration and peak current value was consistent with that observed in PBS, and a good linear relationship was also established (Y = −0.1425X + 1.2143, R2 = 0.9644). Based on the aforementioned relationship, preliminary calculations indicated that limits of detection (LOD) for this sensor in PBS and serum were 1.53 pg mL−1 and 1.74 pg mL−1 respectively. The detection limit was determined using the standard method: the standard deviation (σ) was obtained by measuring the signal values of blank solutions multiple times, and calculated according to the formula LOD = 3.3σ/S (where S is the slope of the calibration curve). To further validate the clinical applicability and accuracy of the sensor, the same clinical samples were analysed using both enzyme-linked immunosorbent assay (ELISA) and the sensor. The results from both methods were shown in Fig. 5C. Experimental results indicated that the sensor's detection outcomes showed no significant divergence from those obtained via ELISA. These findings demonstrated the sensor possesses a favourable linear relationship and a low detection limit. Crucially, the sensor also exhibited sound clinical applicability and accuracy.

Fig. 5. (A) Linear relationship between ALP concentration and DPV peak current value when dissolved in PBS (A) and serum (B) respectively. Insert: DPV detection results for ALP at different concentrations dissolved in PBS and serum. (C) Comparison of two methods for determination of ALP obtained in serum samples.

Fig. 5

Open in a new tab

3.6. Clinical sample testing

To validate the detection capability of the immunosensor in authentic clinical samples, we randomly selected blood samples from ten clinical patients for testing under optimal conditions (Fig. 6A). Experimental results demonstrated that the sensor successfully obtained differentiated response signals from complex serum samples, providing preliminary evidence of its tolerance to matrix interference and its feasibility for clinical testing. Previous studies on staging and kinetics based on OVCF have demonstrated that ALP exhibited a relatively consistent pattern of change: a transient decrease or near-baseline level during the acute phase → followed by an increase → peaking at approximately 30 days → subsequently declining gradually.27 To further evaluate the sensor's applicability in clinical samples, blood samples from different disease progression stages were tested. As shown in Fig. 6B, the results from left to right correspond to: normal healthy state, acute phase of OVCF, post-OVCF surgery, one month post-surgery, and two months post-surgery. To minimise interference from disease progression, only patients undergoing surgery within days of injury were selected. Experimental results indicated that the trend of ALP changes in blood samples from patients at different stages of disease progression aligns with previous studies. This demonstrated that the sensor developed in this study possesses favourable clinical applicability. Consequently, we further examined serum ALP antigen levels in patients undergoing two distinct surgical approaches (unilateral injection (U) and bilateral injection (B)) during the acute phase and one month post-operatively (Fig. 6C). In the acute phase, serum ALP antigen levels appeared comparable between the unilateral and bilateral injection groups. At one month post-surgery, serum ALP antigen levels in the bilateral injection group tended to be higher than those in the unilateral injection group. This suggested that the two injection strategies may be associated with different postoperative ALP antigen responses. However, because the present sensor measured ALP antigen concentration rather than enzymatic activity, this finding should not be directly interpreted as evidence of enhanced ALP catalytic activity or increased mineralisation function. From a surgical perspective, bilateral approaches typically permit greater bone cement injection volumes and more symmetrical vertebral augmentation. Where unilateral injection fails to cross the midline, resulting in unilateral augmentation, a bilateral strategy may be required to improve distribution and stability.4 These mechanical differences may partly explain the observed tendency in postoperative ALP antigen levels, but this interpretation remains speculative because cement volume, baseline bone turnover status, liver and biliary function, and other clinical confounders were not fully controlled. It should be emphasised that serum ALP antigen levels may be influenced by hepatic and biliary function, baseline bone turnover status, fracture severity, cement volume, medication history, and postoperative rehabilitation status. Because this comparison was exploratory and observational, and because multivariable adjustment for these confounders was not performed, the present findings should not be interpreted as evidence that bilateral injection enhances bone repair or mineralisation. Larger prospective studies with predefined statistical analysis and adequate confounder adjustment are required for further validation.4,5

Fig. 6. (A) ALP test results for 10 randomly selected clinical patients. (B) ALP concentration test results at different stages of the disease, from left to right correspond to: normal healthy state, acute phase of OVCF, post-OVCF surgery, one month post-surgery, and two months post-surgery. (C) Comparison of ALP concentration measurements in patients undergoing unilateral (U) versus bilateral (B) injections, both preoperatively (U1, B1) and at one month postoperatively (U2, B2).

Fig. 6

Open in a new tab

3.7. Limitations of the study

A limitation of this study is that the AgNPs–DTNB interfacial interaction was mainly supported by the known sulfur–silver affinity and the stepwise electrochemical characterization of the modified electrode. Direct surface chemical evidence, such as XPS or Raman-based comparison of AgNPs before and after DTNB incubation, was not included because the present work focused on the construction, analytical performance, and clinical applicability of the ALP immunosensor rather than on a dedicated mechanistic investigation of Ag–S bonding states. Future studies will further clarify the detailed surface chemical states of the AgNPs–DTNB interface.

A further limitation is that this study measured serum ALP antigen concentration rather than ALP enzymatic activity. Although ALP is biologically involved in bone mineralisation, antigen abundance and catalytic activity are not equivalent. In the absence of parallel ALP activity assays, bone-specific ALP isoenzyme analysis, or direct mineralisation-related validation, the present clinical results should be interpreted as perioperative changes in ALP antigen levels rather than direct evidence of altered ALP enzymatic activity or mineralisation capacity. In addition, the comparison between unilateral and bilateral injection groups was exploratory and based on a limited clinical sample set. Potential confounders affecting serum ALP antigen levels, including hepatic and biliary function, baseline bone turnover status, fracture severity, bone cement volume, medication history, and postoperative rehabilitation status, were not fully controlled. Therefore, the observed tendency toward higher ALP antigen levels in the bilateral injection group at one month post-surgery should not be interpreted as evidence that bilateral injection enhances bone repair or mineralisation. Larger prospective studies with predefined statistical analysis, adequate sample size, parallel ALP activity assessment, and multivariable adjustment for clinical confounders are required for further validation.

4. Conclusion

In this study, we developed an electrochemical immunosensor for the detection of serum ALP antigen concentration. The sensor showed favourable analytical performance, including stability, reproducibility, specificity, low detection limit, interference resistance, and applicability in serum samples. Clinical sample analysis preliminarily indicated that different PVP injection strategies may be associated with distinct postoperative ALP antigen trajectories. However, this observation should be regarded as exploratory and should not be interpreted as evidence of enhanced bone repair, because the present platform detects ALP antigen rather than enzymatic activity and potential clinical confounders were not fully controlled. Larger prospective studies with parallel ALP activity assessment and appropriate confounder adjustment are needed for further validation.

Ethical statement

All clinical samples were obtained with informed consent signed by the patients. This study strictly adheres to the relevant laws, regulations, and ethical guidelines of the People's Republic of China and has received full approval from the Ethics Committee of the Second Affiliated Hospital of Shandong First Medical University (Ethics approval number: R202106180386).

Author contributions

All authors listed have made a substantial, direct, and intellectual contribution to the work and approved it for publication. Yinfei Xu: revision of the manuscript, response to reviewers' comments, improvement of scientific interpretation, and writing – review and editing. Jiuming Dai: revise the manuscript and reply to the reviewers' comments; Ying Chen: methodology, data analysis, writing – original draft, Maoyuan Bian: collect clinical samples; Cai Wang: writing – review and editing.

Conflicts of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Data availability

The raw data supporting the conclusion of this article will be made available by the authors, without undue reservation.

References

  1. Klazen C. A. H. Lohle P. N. M. de Vries J. Jansen F. H. Tielbeek A. V. Blonk M. C. Venmans A. van Rooij W. J. J. Schoemaker M. C. Juttmann J. R. Lo T. H. Verhaar H. J. J. van der Graaf Y. van Everdingen K. J. Muller A. F. Elgersma O. E. H. Halkema D. R. Fransen H. Janssens X. Buskens E. Mali W. P. T. M. Lancet. 2010;376:1085–1092. doi: 10.1016/S0140-6736(10)60954-3. [DOI] [PubMed] [Google Scholar]
  2. Kallmes D. F. Comstock B. A. Heagerty P. J. Turner J. A. Wilson D. J. Diamond T. H. Edwards R. Gray L. A. Stout L. Owen S. Hollingworth W. Ghdoke B. Annesley-Williams D. J. Ralston S. H. Jarvik J. G. N. Engl. J. Med. 2009;361:569–579. doi: 10.1056/NEJMoa0900563. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Buchbinder R. Osborne R. H. Ebeling P. R. Wark J. D. Mitchell P. Wriedt C. Graves S. Staples M. P. Murphy B. N. Engl. J. Med. 2009;361:557–568. doi: 10.1056/NEJMoa0900429. [DOI] [PubMed] [Google Scholar]
  4. Chen Y.-C. Zhang L. Li E.-N. Ding L.-X. Zhang G.-A. Hou Y. Yuan W. Medicine. 2019;98:8. doi: 10.1097/MD.0000000000014317. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Vimalraj S. Gene. 2020;754:144855. doi: 10.1016/j.gene.2020.144855. [DOI] [PubMed] [Google Scholar]
  6. Orimo H. J. Nippon Med. Sch. 2010;77:4–12. doi: 10.1272/jnms.77.4. [DOI] [PubMed] [Google Scholar]
  7. Schumann G. Aoki R. Ferrero C. A. Ehlers G. Férard G. Gella F. J. Jørgensen P. J. Kanno T. Kessner A. Klauke R. Kytzia H. J. Lessinger J. M. Miller W. G. Nagel R. Pauwels J. Schimmel H. Siekmann L. Weidemann G. Yoshida K. Ceriotti F. Clin. Chem. Lab. Med. 2006;44(9):1146–1155. doi: 10.1515/CCLM.2006.212. [DOI] [PubMed] [Google Scholar]
  8. Zhang L. Nie J. Wang H. Yang J. Wang B. Zhang Y. Li J. Anal. Methods. 2017;9:3375–3379. doi: 10.1039/C7AY00599G. [DOI] [Google Scholar]
  9. Cao F. Wang M. Yi X. Sun D. Analyst. 2022;147:2494–2499. doi: 10.1039/D1AN02159A. [DOI] [PubMed] [Google Scholar]
  10. Yu J. Stanković D. M. Vidic J. Sojic N. Sens. Diagn. 2024;3:1887–1898. doi: 10.1039/D4SD00272E. [DOI] [Google Scholar]
  11. Akyazi T. Basabe-Desmonts L. Benito-Lopez F. Anal. Chim. Acta. 2018;1001:1–17. doi: 10.1016/j.aca.2017.11.010. [DOI] [PubMed] [Google Scholar]
  12. Ellerbee A. K. Phillips S. T. Siegel A. C. Mirica K. A. Martinez A. W. Striehl P. Jain N. Prentiss M. Whitesides G. M. Anal. Chem. 2009;81:8447–8452. doi: 10.1021/ac901307q. [DOI] [PubMed] [Google Scholar]
  13. Yi J. You E.-M. Hu R. Wu D.-Y. Liu G.-K. Yang Z.-L. Zhang H. Gu Y. Wang Y.-H. Wang X. Ma H. Yang Y. Liu J.-Y. Fan F. R. Zhan C. Tian J.-H. Qiao Y. Wang H. Luo S.-H. Meng Z.-D. Mao B.-W. Li J.-F. Ren B. Aizpurua J. Apkarian V. A. Bartlett P. N. Baumberg J. Bell S. E. J. Brolo A. G. Brus L. E. Choo J. Cui L. Deckert V. Domke K. F. Dong Z.-C. Duan S. Faulds K. Frontiera R. Halas N. Haynes C. Itoh T. Kneipp J. Kneipp K. Le Ru E. C. Li Z.-P. Ling X. Y. Lipkowski J. Liz-Marzán L. M. Nam J.-M. Nie S. Nordlander P. Ozaki Y. Panneerselvam R. Popp J. Russell A. E. Schlücker S. Tian Y. Tong L. Xu H. Xu Y. Yang L. Yao J. Zhang J. Zhang Y. Zhang Y. Zhao B. Zenobi R. Schatz G. C. Graham D. Tian Z.-Q. Chem. Soc. Rev. 2025;54:1453–1551. doi: 10.1039/D4CS00883A. [DOI] [PubMed] [Google Scholar]
  14. Sloan-Dennison S. Wallace G. Q. Hassanain W. A. Laing S. Faulds K. Graham D. Nano Converg. 2024;11:33. doi: 10.1186/s40580-024-00443-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Liu Z. Qi W. Xu G. Chem. Soc. Rev. 2015;44:3117–3142. doi: 10.1039/C5CS00086F. [DOI] [PubMed] [Google Scholar]
  16. Giagu G. Fracassa A. Fiorani A. Villani E. Paolucci F. Valenti G. Zanut A. Microchim. Acta. 2024;191:359. doi: 10.1007/s00604-024-06413-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Kim J. Park M. Biosensors. 2021;11:360. doi: 10.3390/bios11100360. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Guo J. Anal. Chem. 2017;89:8609–8613. doi: 10.1021/acs.analchem.7b02531. [DOI] [PubMed] [Google Scholar]
  19. Simão E. P. Frías I. A. M. Andrade C. A. S. Oliveira M. D. L. Colloids Surf., B. 2018;171:413–418. doi: 10.1016/j.colsurfb.2018.07.056. [DOI] [PubMed] [Google Scholar]
  20. Walgama C. Raj N. Chem. Commun. 2023;59:11161–11173. doi: 10.1039/D3CC02561F. [DOI] [PubMed] [Google Scholar]
  21. Zhang Q. Li W. Moran C. Zeng J. Chen J. Wen L.-P. Xia Y. J. Am. Chem. Soc. 2010;132:11372–11378. doi: 10.1021/ja104931h. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Rycenga M. Cobley C. M. Zeng J. Li W. Moran C. H. Zhang Q. Qin D. Xia Y. Chem. Rev. 2011;111:3669–3712. doi: 10.1021/cr100275d. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Chen R. Wang H. Zhao Y. Nan X. Wei W. Du C. Zhang F. Luo Q. Yang L. Xiong B. Nanomaterials. 2022;12:1091. doi: 10.3390/nano12071091. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Katz E. Willner I. Angew. Chem., Int. Ed. 2005;44:4791–4794. doi: 10.1002/anie.200501126. [DOI] [PubMed] [Google Scholar]
  25. Lu W. Qian C. Bi L. Tao L. Ge J. Dong J. Qian W. Biosens. Bioelectron. 2014;53:346–354. doi: 10.1016/j.bios.2013.10.007. [DOI] [PubMed] [Google Scholar]
  26. Zambry N. S. Prathaban S. N. Ibrahim F. Thiha A. Karajacob A. S. Jamaluddin N. F. Omar S. F. S. Shariffuddin I. I. Bustam A. Tay S. T. Microchim. Acta. 2025;192:437. doi: 10.1007/s00604-025-07252-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Qi H. Xue J. Gao J. Zhang Y. Sun J. Wang G. Med. Sci. Monit. 2020;26:346–354. doi: 10.12659/MSM.923713. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The raw data supporting the conclusion of this article will be made available by the authors, without undue reservation.

Articles from RSC Advances are provided here courtesy of Royal Society of Chemistry

RESOURCES