Stemless reverse total shoulder arthroplasty: a systematic review of short- and mid-term results

Abstract

Introduction

Stemless reverse total shoulder arthroplasty is used to treat rotator cuff deficient arthropathies, rheumatoid arthritis, and osteoarthritis. It has several advantages over the stemmed implant including preservation of bone stock, reduced surgical time, and easier revision.

Methods

A systematic search was conducted in MEDLINE, EMBASE, PubMed, and CENTRAL to retrieve all relevant studies evaluating stemless reverse total shoulder arthroplasty.

Results

The literature search identified 1993 studies out of which 7 studies were included in this review; 324 patients underwent stemless reverse total shoulder arthroplasty with a weighted mean age of 74.1 (SD = 8.6, range = 38 to 93) years and a weighted mean follow-up time of 44 (SD = 6.6, range = 3 to 95) months. The included studies reported significant improvements in range of motion and functional scores comparable to stemmed reverse total shoulder arthroplasty. The weight mean flexion and abduction was (135 ± 12)° and (131 ± 12)° post-operatively, respectively. The weighted mean constant score increased from (26.7 ± 5.2) Patients (pts) to (63.0 ± 8.0) pts post-operatively. Overall complication and revision rate were 12.3% and 5.2%.

Conclusion

Early and mid-term results indicate stemless reverse total shoulder arthroplasty has similar clinical outcomes to stemmed reverse total shoulder arthroplasty. There was no radiological evidence of humeral loosening at the latest follow-up.

Background

Reverse total shoulder arthroplasty (rTSA) is increasingly utilized to treat rotator cuff deficient glenohumeral arthropathies, glenoid deformity due to primary osteoarthritis, proximal humeral fractures, rheumatoid arthritis, chronic glenohumeral dislocation, and failed anatomic shoulder replacement. ¹ Most rTSA prostheses are designed with humeral fitting stem. However, implants are evolving rapidly, and with an increased focus on bone preservation, short stem and stemless prostheses have been increasingly utilized. Such implants aim to achieve metaphyseal fixation while minimizing bone resection to better preserve the native bone stock compared to traditional stemmed rTSA prostheses. Preservation of bone stock allows for potentially easier revision surgery, if needed. There is also a lower risk of periprosthetic bone fractures as stemless designs do not require broaching and they also reduce stress shielding by the implant. ² Moreover, stemless implants allow for flexibility with respect to alignment particularly when the humeral diaphysis is altered or in cases of fracture sequalae and malunion. ¹

While there are numerous advantages associated with stemless implants, there are potential risks due to the constrained nature of the implant and increased forces on the humeral component. Such risks include inadequate fixation resulting in early failure as well as reported risk of intraoperative lateral cortex fractures. It is also more difficult to control inclination of the stemless implant which may be important, especially for rTSA. ³ Numerous studies have reported on stemless rTSA since its invention more a decade ago.^4–6 However, no previous systematic reviews have assessed the reported outcomes of stemless rTSA in the literature.

This systematic review includes all early and mid-term results of stemless rTSA and aims to assess clinical and radiological outcomes of stemless rTSA including range of motion (ROM), functional scores, operative time, blood loss, complication rates, revision rates, and radiological loosening.

Methods

Search strategy and eligibility

A search was performed in Ovid Medline (1946 to 30 June 2020), EMBASE (1974 to 29 June 2020), CENTRAL (up to 30 June 2020), and PubMed (up to 30 June 2020) using the keywords “shoulder,” “arthroplasty,” and “stem*” limited to humans and reports in English (online Appendix I).

The inclusion criteria were (1) studies reporting clinical and/or radiological outcomes after stemless rTSA, (2) studies published in English, and (3) studies on humans. The exclusion criteria were (1) non-surgical studies (e.g. review articles, cadaver studies, technique articles), (2) study outcomes not reported separately for stemless rTSA and other surgical procedures (e.g. stemmed rTSA), and (3) radiological studies with no clinical outcomes. For studies that reported on the same cohort, we planned on including the more recent report. For studies with incomplete overlap in cohort, the larger cohort was included.

Study selection

We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines to systematically screen studies. ⁷ Independent reviewers (EL, DK, NY) screened the title and abstract of studies identified in the literature search in duplicate. Relevant studies were then rescreened based on the full-text article in duplicate. Any disagreements at the title and abstract stage or the full-test review stage was resolved either by consensus or through discussion with a third reviewer (NH). References of included articles were hand-searched to identify additional articles.

Data collection

Three independent reviewers extracted data from the included studies in duplicate (EL, DK, NY). Disagreements at the data extraction stage were resolved either by consensus or through discussion with a third reviewer (NH). The data collected included study characteristics (e.g. author, study title, year of publication, study design, level of evidence), patient demographics (e.g. age, sex), surgery information (e.g. technique, implant, indication), clinical outcomes (e.g. ROM, functional scores, patient satisfaction), radiological outcomes (e.g. radiolucencies, loosening, scapular notching), and complications (e.g. type of complication, management, revisions). We also recorded information about conflicts of interest and sources of funding.

Quality assessment of included studies

Independent reviewers (EL, DK, NY) performed quality assessment of the included studies in duplicate using the Methodological Index for Non-randomized Studies (MINORS) appraisal tool for observational studies. ⁸ Disagreements at the quality assessment stage were resolved either by consensus or through discussion with a third reviewer (NH).

The MINORS appraisal tool was used to evaluate the quality of observational studies. ⁸ A score of 0, 1, or 2 was assigned to each of the 12 criteria on the MINORS checklist, resulting in a maximum score of 16 for non-comparative studies and 24 for comparative studies. ⁸ MINORS scores were categorized as follows: 0–6 to indicate very low-quality evidence; 7–10 to indicate low quality of evidence; 10–16 to indicate fair quality of evidence; and >16 to indicate good quality of evidence.

Statistical analysis

Descriptive statistics

Mean, standard deviations (SD), confidence intervals (CI), and range were calculated and presented when applicable. Weighted mean and weighted standard deviations, which takes consideration of the different sample sizes of studies, were calculated and presented when appropriate. Follow-up time was categorized as follows: short-term (0–2 years); mid-term (2–10 years); and long-term (>10 years).

Inter-rater agreement

A kappa (κ) statistic indicating inter-reviewer agreement was calculated for all screening stages and categorized as follows: 0.81–0.99 to indicate excellent agreement; 0.61 to 0.80 to indicate substantial agreement; 0.41 to 0.60 to indicate moderate agreement; 0.21 to 0.40 to indicate fair agreement; and 0.20 or less to indicate slight agreement. ⁹

Results

Included studies

The literature search identified 1993 studies out of which 7 studies were included in this systematic review following abstract and full-text review (Figure 1). Five studies were excluded due to outcomes reported on the same patient cohort.^10–14 Three reviewers screened the papers in duplicate. There was substantial agreement between reviewers (DK, EL) at the title and abstract stage (κ = 0.702, 95% CI = 0.603–0.800) and moderate agreement at the full-text screening stage (κ = 0.560, 95% CI = 0.292–0.827). There was substantial agreement between reviewers (DK, NY) at the title and abstract stage (κ = 0.782, 95% CI = 0.688–0.875) and excellent agreement at the full-text screening stage (κ = 0.881, 95% CI = 0.650–1.000).

Figure 1.

PRISMA flow diagram.

Demographics

This systematic review included 324 patients who underwent stemless rTSA. The mean sample size per study was 46 (SD = 35.3, range = 7 to 98) for stemless patients. The weighted mean age was 74.1 (SD = 8.6, range = 38 to 93) years; 125 (39%) and 199 (61%) of the included patients were male and female, respectively. The studies were conducted in the following countries including Austria (n = 24), France (n = 143), Italy (n = 24), Sweden (n = 16), and the United Kingdom (n = 117) (Table 1).

Table 1.

Characteristics of included studies.

Primary author	Year	Study design (level of evidence)	Country	Dates of study	No. stemless rTSA patients	No. Female (%)	Mean follow-up (range)	MINORS score
Ballas and Béguin 4	2013	Prospective case series (IV)	France	2004–2009	56	40 (71)	59 (38–95) months	13/16
Kadum et al. 5	2013	Prospective case series (IV)	Sweden	2007–2012	16	6 (38)	39 (15–66) months	19/24
Leonidou 17	2019	Prospective case series (IV)	UK	2009–2016	36	27 (75)	3 (1–7) years	8/16
Levy et al. 15	2016	Prospective case series (IV)	UK	2005–2010	98	78 (80)	50 (24–82) months	14/16
Micheloni et al. 18	2019	Case series (IV)	Italy	2016–2018	7	5 (71)	5.5 (3–9) months	8/16
Moroder et al. 6	2016	Case control (III)	Austria	2009–2013	24	17 (71)	34.2 (NR) months	18/24
Teissier et al. 16	2015	Prospective case series (IV)	France	2006–2010	87	26 (30)	41 (24–69) months	9/16

NR: not reported; rTSA: reverse total shoulder arthroplasty; MINORS: Methodological Index for Non-randomized Studies.

Follow-up

The weighted mean follow-up time was 44 (SD = 6.6, Range = 3 to 95) months. There was one short-term (0–2 years follow-up) study, six mid-term (2–10 years follow-up) studies, and no long-term (>10 years follow-up) studies. There was substantial loss to follow-up mainly due to unrelated deaths, decline to follow-up due to unrelated reasons, and patients moving away.

Competing Interests

Out of the included studies, four studies ^4,6,15,16 had conflicts of interests where the authors received either funding or consultant payments from the manufacturers of the implants. One study ¹⁷ reported no conflict of interest. The remaining two studies^5,18 did not report on conflicts of interest.

Quality assessment

All of the included studies were non-randomized in design and their quality was assessed using the MINORS score. Six (86%) of the included studies were level IV and one (14%) of the included studies was level III. The non-randomized studies can be further divided into non-comparative and comparative studies. The mean MINORS score for non-comparative studies was 11.0 ± 2.9 out of 16 (4 studies), indicating that the non-comparative studies were fair in quality. The mean MINORS score for comparative studies was 17.0 ± 2.64 out of 24 (3 studies), indicating that the comparative studies were good in quality (Table 1).

Surgical information

Prosthesis

Four studies^4–6,16 (183 patients) reported on the TESS (Biomet, Warsaw, USA) stemless rTSA implant and three studies^15,17,18 (141 patients) reported on the Verso (Innovative Design Orthopaedics, London, UK) stemless rTSA.

Two studies reported on the size of the glenosphere. Leonidou et al. reported that the glenosphere was available in two diameters (36 mm and 41 mm) for the Verso prosthesis. ¹⁷ Teissier et al. reported that the glenosphere was available in two diameters (36 mm and 41 mm) for the TESS prosthesis. ¹⁶

Four studies reported on the resection of the humeral bone. Ballas and Béguin reported that a cutting guide stem was introduced along the axis of the humeral shaft and a metaphyseal bone cut was performed with the help of a 150° tilt guide. ⁴ Micheloni et al. reported that minimal humeral bone was resected cancellous bone was used for bone graft for further stability. Levy et al. reported that a humeral cut was made at a 155° angle, with the final implant angle of 145° using the inclined liner. ¹⁵ Teissier et al. reported that the humerus was cut with a 150° guide and the short reverse corolla was inserted at 20° retroverted. ¹⁶

Indications

Indication for surgery included rotator cuff tear and/or arthropathy in 271 (82%) patients, rheumatoid arthritis in 22 patients (7%), proximal humerus fracture sequelae in 15 patients (5%), primary osteoarthritis in 8 patients (2%), failed prosthesis in 4 (1%) patients, trauma in 3 patients (1%), and dislocations in 2 patients (1%) (Table 2).

Table 2.

Indication for stemless rTSA.

Indication	No. of patients (%)
Rotator cuff tear and/or arthropathy	271 (81)
Rheumatoid arthritis	22 (7)
Proximal humerus fracture sequelae	15 (5)
Primary osteoarthritis	8 (2)
Failed prosthesis	4 (1)
Trauma	3 (1)
Dislocation	2 (1)

Four studies reported on Hamada and Fukuda classification of the rotator cuff tears. Ballas and Béguin reported that 46 shoulders (82%) were Hamada and Fukuda I, II, or III and 10 shoulders (18%) were grade IV or V. ⁴ Levy et al. reported that 2 (2%) patients were Hamada and Fukuda grade II, 8 (8%) were grade III, 18 (18%) were grade IVa, 34 (35%) were grade IVb, and 40 (41%) were grade V. ¹⁵ Moroder et al. reported that 5 (21%) patients were Hamada and Fukuda grade II, 8 (33%) patients were grade III, 7 (29%) patients were grade IV, and 4 (17%) patients were grade V. ⁶ Teissier et al. reported that 45% of patients were Hamada and Fukuda stage III, and 46% of patients were stage IV. ¹⁶ Additionally, Kadum et al. reported that rotator cuff dysfunction is defined as an irreparable rotator cuff found intraoperatively.

Clinical outcomes

Range of motion

Stemless rTSA resulted in significant improvements in ROM for all included studies. The weighted mean flexion was (70 ± 8°, n = 264) pre-operatively and (135 ± 12°, n = 300) post-operatively. The weighted mean abduction was (78 ± 9°, n = 110) pre-operatively and (131 ± 12°, n = 146) post-operatively. There was an inconsistent use of different measurement systems for internal and external rotation including point systems, degrees, and body parts. Nevertheless, all authors reported significant improvements in internal and external rotation after the procedure. No significant differences in ROM were reported between stemless and stemmed implants in the two comparative studies.^5,15 One comparative study did not comment on the significance in ROM between stemless and stemmed implants ¹⁸ (Table 3).

Table 3.

Range of motion after stemless rTSA.

Study	No. stemless rTSA patients	Flexion	Abduction	Internal rotation	External rotation
Ballas and Béguin 4	56	Pre: 79° (20–150) Post: 140° (90–170)	NR	NR	Pre: 13° (−40–70) Post: 45° (0–70)
Kadum et al. 5	16	Pre: 50° (10–80) Post: 110° (80–170)	Pre: 30° (10–60) Post: 110° (60–170)	Sacrum (trochanter L5)	L3 (trochanter L2)
Leonidou et al. 17	36	Post: 133° (60–160)	Post: 110° (60–160)	NR	Post: 30° (10–60)
Levy et al. 15	98	Pre: 47° Post:129°	NR	Pre: 21° Post: 65°	Pre: 10° Post: 51°
Micheloni et al. 18	7	Pre: 40.7° (30–60) Post: 154.3° (120–180)	Pre: 52.8° (45–70) Post: 150° (110–180)	Pre: 17.9° (10–30) Post: 65.7° (50–90)	Pre: 3.57° (0–10) Post: 51.4° (15–75)
Moroder et al. 6	24	Post: 7.8 ± 1.9 pts	Post: 6.9 ± 2.0 pts	Post:5.3 ± 2.3 pts	Post: 6.6 ± 2.6 pts
Teissier et al. 16	87	Pre: 96° (0–160) Post: 143° (90–170)	Pre: 89° (0–160) Post: 138° (80–160)	Pre: 5° (2–10) Post: 4° (2–8)	Pre:26° (−60 – 70) Post: 39° (20–70)

NR: not reported; rTSA: reverse total shoulder arthroplasty.

Functional scores

Stemless rTSA resulted in significant improvements in functional scores in all included studies. Different functional scales were used between studies which makes it difficult to compare results between studies. The constant score was the most consistently used functional score. The weighted mean constant score was (27 ± 5 Patients (pts), n = 248) pre-operatively and (63 ± 8 pts, n = 308) post-operatively. No significant differences in constant score were noted between stemless and stemmed groups in the two comparative studies.^5,15 One comparative study did not comment on the significance in functional scores between stemless and stemmed implants ¹⁸ (Table 4).

Table 4.

Functional scores after stemless rTSA.

Study	Constant score	Adjusted constant score	ASES score	Quick DASH score	Oxford shoulder score	ADLEIR	Simple shoulder test	Subjective shoulder value
Ballas and Béguin 4	Pre: 29 (16–59; SD = 8) Post: 62 (38–85; SD = 12)	NR	NR	NR	Pre: 46 (34–65; SD = 5) Post: 17 (12–30; SD = 4)	NR	NR	NR
Kadum et al. 5	NR	NR	NR	Pre: 67 (38.6–88.6) Post: 29 (4.5–86.4)	NR	NR	NR	NR
Leonidou 17	Post: 63 (35–86)	NR	NR	NR	Pre: 11 (2–19) Post: 44 (29–48)	Pre: 12 (0–27) Post: 31 (18–36)	NR	NR
Levy et al. 15	Pre: 14 Post: 59	Pre: 21 Post: 86	NR	NR	NR	NR	NR	Pre:8/100 Post: 85/100
Micheloni et al. 18	Pre: 21.6 (SD = 4.5) Post: 56.9 (SD = 12.3)	NR	NR	NR	NR	NR	Pre: 2.3 (SD = 1.1) Post: 9.4 (SD = 1.5)	NR
Moroder et al. 6	Post: 65.4 (SD = 12.9)	NR	Post: 76.2 (SD = 10.8)	NR	NR	NR	NR	Post: 86.6 ± 11.9
Teissier et al. 16	Pre: 40 (12–76; SD = 24) Post: 68 (42–100; SD = 12)	Post: 90	NR	NR	NR	NR	NR	NR

ADLEIR: Activities of daily living with requirement for external rotation and internal rotation; ASES: American Shoulder and Elbow Surgeons; DASH: Disabilities of the Arm, Shoulder and Hand; NR: not reported.

Pain

Five of the included studies reported on pain. There was significant improvement in post-operative pain in all included studies (Table 5).

Table 5.

Pain scores after stemless rTSA.

Study	Pain VAS
Kadum et al. 5	Pre: 30 Post: 10
Levy et al. 15	Pre: 12/15 Post: 2/15
Micheloni et al. 18	Pre: 14.3/15 Post: 4.9/15
Moroder et al. 6	Post: 0.4
Teissier et al. 16	Pre: 8/15 Post: 2/15

NR: not reported; VAS: Visual Analogue Scale.

Complications and revisions

There were 40 (12%) reported complications including 17 periprosthetic fractures (5%), 6 glenoid loosening (2%), 4 dislocation (1%), 3 early humeral component displacement (1%), 3 post-operative stiffness (1%), 3 glenoid dissociation (1%), 2 infections (<1%), 1 hematoma (<1%), and 1 ruptured subscapularis (<1%) (online Appendix II).

There were a total of 17 (5%) reported revisions including 4 for dislocations (1%), 3 for glenoid loosening (1%), 3 for early humeral component displacement (1%), 3 for glenoid dissociation (1%), 3 for periprosthetic fracture (1%), and 1 for implant removal due to poor health status not further specified (<1%) (online Appendix III).

Radiological outcomes

Radiological outcomes were reported for all included studies. There were 6 (2%) cases of glenoid loosening and no cases of late humeral loosening. There were no cases of humeral radiolucency, 53 (16%) cases of glenoid radiolucency, and 3 (1%) cases of periprosthetic radiolucency not further specified. There were 56 (17%) cases of scapular notching.

Two studies reported on humeral component neck-shaft angle (NSA) on radiological follow-up.^6,16 Moroder et al. (6 reported the NSA to be 134° (range = 116–152°). Teissier et al. ¹⁶ reported the NSA to be 154° (range = 142–165°). There was no significant association between NSA and scapular notching (p = 0.18). ¹⁶

One study reported on glenoid overhang. Kadum et al. reported the mean glenoid overhang to be 1.3 mm and that there was higher incidence of SN with no overhang (p < 0.001). ⁵

Other outcomes

Moroder et al. ⁶ found surgical times to be significantly shorter in the stemless group (81 minutes vs. 110 minutes, p < 0.001) compared to the stemmed group. Two patients in the stemmed group required blood transfusions, and none in the stemless group required blood transfusions though the difference is not significant (p = 0.180).

Discussion

Stemless rTSA has several proposed advantages over stemmed rTSA including reduced operation times, improved bone preservation, and potentially easier revision. As the popularity of stemless rTSA continues, critical assessment of outcomes is essential given current concerns with implant fixation. This review included all rTSA prosthesis that have been described as stemless in published literature and were approved as stemless implants.

The most important finding of this review was similar complication rates to those reported from stemmed rTSA. The overall complication rate in this review was 12% with the most common complication being periprosthetic fractures (5%), glenoid loosening (2%), dislocation (1%), and early humeral component displacement (1%). The overall revision rate was 5% with the most common causes being dislocations (1%), glenoid loosening (1%), and early humeral component displacement (1%). There were no cases of late humeral loosening reported in any of the included studies.

A systematic review conducted by Zumstein et al. found complication rates and revision rates in stemmed rTSA to be 21% and 10%, respectively in 782 patients. ¹⁹ The most common post-operative complications included instability (5%), infection (4%), and glenoid loosening (4%). The mean follow-up time in the review by Zumstein et al. was 42 months which is comparable to our weighted mean follow up time of 44 (SD = 6.6, Range = 3 to 95) months. However, Zumstein et al. only included patients with minimum 24 months follow-up whereas the present review included studies with shorter follow-up time which could explain the higher complication and revision rates in their series. ¹⁹

Both stemmed and stemless rTSA results in significant improvements in function. This review found stemless rTSA resulted in significant improvements in ROM and functional scores in all of the included studies. The weighted mean constant score was (27 ± 5) pts pre-operatively and (63 ± 8) pts post-operatively. The weight mean flexion and abduction increased by 65° and 53°, respectively. These improvements exceed the minimal clinically significant difference report by Simovitch et al. for shoulder arthroplasty. ²⁰ There were no significant differences between stemless and stemmed rTSA in ROM and constant scores.

One contraindication to the stemless rTSA implant reported in the literature is advanced osteopenia. Humeral component displacement is an early complication that was encountered in two of the included studies in three cases.^4,5 One case was due to reduction of a postoperative dislocation and two cases were due to inadequate bone quality and fixation which were subsequently revised to a stemmed prosthesis. Kadum et al. ⁵ commented in their study that the stemless implant might not be appropriate in fragility fractures where there would be inadequate fixation due to poor bone quality. No early humeral component failure was reported in the anatomical stemless TSA systematic review. ²¹ Several factors could account for this difference including the constrained nature of the reversed stemless implant, increased forces on the humeral component, and the potential for bony impingement. Nevertheless, these are early results from a limited number of studies and more information is needed on the performance of stemless implants in poor bone quality cases.

An improved method to objectively assess bone quality is required to better identify patients suitable for stemless rTSA. Teissier et al. ¹⁶ utilized a three-step intraoperative test as selection criteria for the stemless rTSA implant that included (1) intact cortical ring after performing the humeral head osteotomy, (2) sufficient resistance of the trabeculae of the osteotomy side to pressure applied by the surgeon’s thumb (thumb test), and (3) no visible cysts at the osteotomy site. These criteria could potentially be used by other investigators in the future.

There was limited reporting on the outcomes that could potentially differentiate stemless and stemmed rTSA including operative times and blood loss. Only one study ⁶ reported on operative times, blood loss, and hospital length of stay. The study found operative times to be significantly shorter (81 min vs. 110 min, p < 0.001) in the stemless group and blood transfusion to be lower (0 patients vs. 2 patients, p = 0.180) but the difference was not significant. Similarly, a previous meta-analysis on stemless anatomical TSA found operative times and blood loss to be significantly shorter and lower in the stemless group. ²¹

Strengths and limitations

This review’s strength resides in its rigorous methodology. Multiple databases were screened, and included references were hand searched for additional articles. Abstract screening, full-text screening, data extraction, and quality assessment were all completed in duplicate by blinded reviewers with conflicts resolved through consensus or discussion with a third reviewer.

Limitations of this review include smaller sample size as there are not that many studies available on stemless rTSA. Additionally, many of the existing studies were sponsored by the manufacturer of the implant. The studies were all conducted in Europe with limited geographic variety. In many of the included studies, the surgeons performing the procedure were consultants for the company designing the prosthesis, and therefore, a potential expertise bias may be present. Outcomes may be less favorable for surgeons who are less familiar with the implant design. Furthermore, the designs of all components in the study were inlay designs. These results cannot be extrapolated to onlay designs which may have higher risk of failure due to increased forces applied with humeral lateralization. Finally, most of the included studies were non-comparative which made it difficult to assess the effectiveness of stemless TSA compared to other procedures because there is no control group.

Conclusion

Early and mid-term results indicate stemless rTSA has similar clinical outcomes to stemmed rTSA. Early evidence indicates that operative time might be lower in stemless compared to stemmed rTSA but more studies are needed to validate this claim. No evidence of humeral loosening is seen at latest radiological follow-up. Surgeons should be cautious of using stemless implants in cases of severe osteopenia due to inadequate humeral fixation which can lead to displacement of the humeral component requiring revision.

ORCID iDs

Bashar Alolabi https://orcid.org/0000-0002-8393-0953

Moin Khan https://orcid.org/0000-0002-8237-8095