Skip to main content
Annals of African Medicine logoLink to Annals of African Medicine
. 2025 Nov 26;25(5):1182–1186. doi: 10.4103/aam.aam_304_25

Investigating the Magnetic Resonance Imaging Cross-sectional Area That Best Correlates with Intraoperative Hamstring Autograft Size

Govind Kumar Gupta 1, Anil Kumar Das 1,✉, Vinay Prabhat 1, Sudha Rani 1, Ratnajeet Chakraborty 2
PMCID: PMC13498331  PMID: 41288395

Abstract

Background:

Hamstring tendon autografts are widely used for anterior cruciate ligament (ACL) reconstruction, but preoperative prediction of graft diameter remains a clinical challenge. Magnetic resonance imaging (MRI)-based cross-sectional area (CSA) measurements are increasingly being used to estimate graft adequacy.

Objective:

To determine which MRI-based CSA measurement level most accurately correlates with intraoperative hamstring autograft diameter in ACL reconstruction.

Methodology:

A prospective observational study was conducted on 41 patients aged 20–45 years undergoing ACL reconstruction. MRI CSA of the semitendinosus and gracilis tendons was measured at two anatomical levels: anteroposterior (AP) and joint line. Correlation with intraoperative graft size was assessed using Pearson and Spearman coefficients. MRI data were analyzed using Horos, and statistical analysis was performed in JASP software.

Results:

The mean CSA at the AP level (19.63 mm²) was higher than at the joint line (16.55 mm²). The AP CSA showed the strongest correlation with intraoperative graft diameter (r = 0.858, P < 0.001), followed by joint line CSA (r = 0.625). The average intraoperative graft diameter was 8.48 mm.

Conclusion:

MRI-based CSA measurement at the AP level has proven to be a reliable, noninvasive predictor of intraoperative hamstring graft diameter, aiding in surgical planning and graft sufficiency prediction.

Keywords: Anterior cruciate ligament reconstruction, cross-sectional area, graft prediction, hamstring autograft, magnetic resonance imaging

INTRODUCTION

Anterior cruciate ligament (ACL) reconstruction commonly uses hamstring tendon autografts, especially within orthopedic sports medicine as a performed procedure. The harvested autograft’s diameter is a key factor for ACL reconstructions’ success and longevity because grafts <8 mm correlate with a higher failure rate.[1] However, it is still a challenge for one to predict graft size. Before operating, graft size estimation is still a challenge. The cross-sectional area (CSA) of the semitendinosus and gracilis tendons was measured by two independent observers at two anatomical levels on MRI. Intraoperatively, the diameters of the harvested hamstring tendons were measured and recorded. These intraoperative graft diameters were then compared and correlated with the preoperatively measured tendon CSA on MRI.

Several studies have examined the association between MRI-derived cross-sectional areas (CSA) of the semitendinosus and gracilis tendons and intraoperative graft size. Hollnagel et al.[1] demonstrated a strong correlation between MRI measurements taken at the level of the medial femoral condyle and intraoperative graft diameter. Their findings suggested that CSA values ranging from 17.5 to 18.8 mm² reliably predicted graft diameters of 8 mm or more. Similarly, Bickel et al.[2] were among the first to report that a combined CSA of ≥18 mm² was associated with an 88% probability of obtaining a graft of adequate size, supporting the role of MRI as a useful preoperative predictive tool.

A 2024 systematic review involving 4140 patients confirmed that MRI-based CSA measurements of the semitendinosus and gracilis tendons are the most consistent predictors of graft adequacy, outperforming anthropometric parameters. Over 60% of the studies reviewed demonstrated moderate-to-strong correlations between the combined CSA of these tendons and intraoperative graft diameter. However, reported cutoff values varied significantly, ranging from 15.8 mm² to 31.2 mm², highlighting the need to standardize both the MRI slice level and measurement technique.[3]

Recent research has aimed to determine the optimal anatomical level for CSA measurement. One study identified the medial epicondyle of the distal femur as the most reliable site, showing the highest correlation with intraoperative graft diameter (r = 0.389).[4] Ayres et al.[5] proposed a regression model using preoperative CSA values to estimate five-stranded hamstring graft size: Graft Diameter (mm) = 5.324 + 0.124 × CSA (st) +0.183 × CSA (gr).[5]

An alternative approach introduced by Kocazeybek et al.[6] utilized the circle area formula to convert total tendon CSA into an estimated graft diameter. This method showed a significant correlation (r = 0.51) with intraoperative measurements and demonstrated better predictive accuracy when CSA was measured at the joint line rather than at the medial femoral condyle.

Image resolution has also been shown to influence MRI predictive accuracy. Erquicia et al.[7] found that × 4 at the widest point of the femoral epicondyle provided a much stronger correlation (r = 0.86) compared to × 2 (r = 0.536). Using this technique, a CSA threshold of 17 mm² consistently predicted graft diameters of ≥8 mm.

In contrast, anthropometric factors such as height and weight showed weaker correlations. Wotherspoon et al.[8] reported moderate associations (r = 0.52 and 0.50, respectively), but MRI-based CSA measurements had significantly stronger predictive value, especially when the two modalities were combined.

Further supporting the utility of MRI, Yan et al.[9] reported high correlation coefficients of 0.858 for semitendinosus, 0.728 for gracilis, and 0.842 for combined CSA, using an 18 mm² threshold to reliably predict grafts ≥8 mm in diameter.

In resource-limited settings, MRI may still offer reliable predictive value. Shrestha et al.[10] demonstrated that a CSA cutoff of 13.3 mm² predicted 7 mm grafts with 85.7% sensitivity and specificity in a cohort of 42 patients, reinforcing the utility of MRI where graft adequacy is critical.

Finally, Hollnagel et al. and Wernecke et al.[11] highlighted the correlation between preoperative combined CSA values and double-bundle graft diameters. Their findings recommended individual thresholds of 10 mm² for gracilis and 17 mm² for semitendinosus tendons for achieving adequate graft sizes during reconstruction.

METHODOLOGY

Study design

This was a prospective observational comparative study conducted to evaluate the correlation between preoperative MRI-measured CSA of hamstring tendons and intraoperative graft diameter in ACL reconstruction. The study involved no interventions and followed standard clinical protocols.

Study setting

The study was carried out at the department of orthopedics. MRI scans were performed in the radiology department, and surgeries were performed in the orthopedic operation theater. Collaboration between departments ensured accurate data collection.

Study duration

The study was conducted for 1 year following ethical approval. This duration allowed for patient enrollment, imaging, surgery, and data analysis within a consistent clinical timeline.

Participants – inclusion and exclusion

Inclusion criteria were patients aged 20–45 years with chronic ACL injuries, with or without meniscal tears. Exclusion criteria included acute injuries, prior surgeries, knee infections, deformities, degenerative changes, metabolic or inflammatory disorders, and multiligament injuries.

Study sampling

Consecutive sampling was used to recruit all eligible patients visiting the orthopedic outpatient department. Patients were enrolled after screening and consent, ensuring a representative clinical sample without selection bias.

Study sample size

Based on statistical calculations for correlation studies, a sample size of 41 was estimated. To allow for attrition, the final target sample was set at 50 patients to maintain study power.

Study groups

No separate groups were formed. Patients were internally compared based on CSA measurements taken at two anatomical MRI levels – the joint line and the medial femoral condyle.

Study parameters

Primary parameter was the MRI-measured CSA of the semitendinosus and gracilis tendons. Secondary parameters included tendon length and intraoperative graft diameter, assessed for correlation and reliability.

Study procedure

After obtaining informed consent, patients underwent MRI scans. The cross-sectional area (CSA) of the semitendinosus and gracilis tendons was measured at two anatomical levels by two independent observers. During surgery, the hamstring tendons were harvested, prepared, and the graft diameter was measured for correlation with the MRI findings.

Study data collection

MRI measurements were recorded using Horos software. CSA values were documented by two observers at two time points. Intraoperative measurements were taken immediately postharvest to avoid variability.

Data analysis

Data were analyzed using JASP software. Pearson and Spearman correlation tests assessed measurement reliability and correlation. Linear regression was used to create predictive models (P ≤ 0.05).

Ethical considerations

Ethical approval was obtained from the Institutional Ethics Committee, RIMS, Ranchi. (Approval No. RIMS/IEC/2023/47, dated June 15, 2024).

RESULTS

Age distribution

The average age of the study participants was 34.39 years, indicating a young adult population. The narrow age range suggests homogeneity in age-related tendon properties [Table 1].

Table 1.

Descriptive statistics – age distribution

Parameter Value
Mode 37.00
Mean 34.390
Standard Deviation 5.375
Range 21.00
Minimum 23.00
Maximum 44.00

Gender distribution

The study cohort was predominantly male (73.17%), which may influence tendon characteristics and graft size. This gender skew reflects the typical ACL injury demographics [Table 2].

Table 2.

Genderwise distribution

Gender Frequency (n) Percentage (%)
Male 30 73.17%
Female 11 26.83%

Descriptive statistics of combined cross-sectional area – magnetic resonance imaging

The mean CSA at the anteroposterior (AP) level (19.63 mm²) was higher than the joint line level (16.55 mm²), suggesting the AP level may be more predictive of graft size [Table 3].

Table 3.

Magnetic resonance imaging cross-sectional area measurements

Parameter Cross Section AP Cross Section Joint Line
Mode 18.9 15.4
Mean 19.627 16.554
Standard Deviation 1.888 1.885
Range 7.220 7.504
Minimum 15.680 13.498
Maximum 22.900 21.002

Intraoperative graft diameter

The mean graft diameter was 8.48 mm, with most grafts measuring 9 mm, indicating clinically adequate graft size in most cases [Table 4].

Table 4.

Intraoperative graft diameter measurements

Parameter Value
Mode 9.00
Mean 8.476
Standard Deviation 0.866
Range 3.00
Minimum 7.00
Maximum 10.00

Correlation Between magnetic resonance imaging cross-sectional area and graft size

Cross-sectional AP showed the strongest correlation with intraoperative diameter (r = 0.858), making it the most reliable MRI predictor [Table 5].

Table 5.

Correlation of magnetic resonance imaging cross-sectional area and intraoperative graft size

Variable Cross section ap Cross section joint line Intraoperative measurement
Cross section AP
 Pearson’s r -
 P -
 Spearman’s rho -
 P -
Cross section joint line
 Pearson’s r 0.756 -
 P <.001 -
 Spearman’s rho 0.723 -
 P <.001
Intraoperative measurement
 Pearson’s r 0.858 0.625
 P <.001 <.001
 Spearman’s rho 0.842 0.610
 P <.001 <.001

DISCUSSION

This study aimed to identify the MRI CSA level that best correlates with intraoperative hamstring autograft diameter – a key factor in the success and longevity of ACL reconstruction. Our findings show that CSA measured at the AP level had a significantly stronger correlation with intraoperative graft diameter compared to measurements taken at the joint line, with a Pearson correlation coefficient of 0.858 (P < 0.001).

These results are in line with earlier research. Erquicia et al.[12] found that MRI CSA measurements at the level of the medial epicondyle correlated well with intraoperative graft diameter, although with a slightly lower correlation than observed in our study. Likewise, Yasumoto et al.[13] reported a moderate correlation (r = 0.73) between tendon CSA and graft diameter. The stronger correlation seen in our study may be due to our choice of measurement level; the AP level may provide better visualization and reduced anatomical variability.

The demographic profile of our study population – primarily young adult males – is consistent with the known epidemiology of ACL injuries. This homogeneity may have contributed to the consistency in tendon morphology and the strong correlations observed.

Our mean graft diameter of 8.48 mm exceeds the threshold commonly associated with reduced failure rates. These results support the use of MRI CSA measurement at the AP level as a valuable, noninvasive preoperative tool to estimate graft adequacy. Incorporating this approach into surgical planning can improve graft selection and reduce intraoperative uncertainty, particularly in resource-limited settings.

CONCLUSION

This study concludes that the MRI-measured CSA of hamstring tendons at the AP level is a highly accurate predictor of intraoperative graft size in ACL reconstruction. With a strong correlation coefficient (r = 0.858), it offers a valuable preoperative assessment tool, especially in resource-limited settings. Implementing this imaging approach can enhance surgical planning, reduce intraoperative uncertainty, and ensure graft adequacy, ultimately improving clinical outcomes and long-term graft success in ACL reconstructions.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.

REFERENCES

  • 1.Hollnagel K, Johnson BM, Whitmer KK, Hanna A, Miller TK. Prediction of autograft hamstring size for anterior cruciate ligament reconstruction using MRI. Clin Orthop Relat Res. 2019;477:2677–84. doi: 10.1097/CORR.0000000000000952. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Bickel BA, Fowler TT, Mowbray JG, Adler B, Klingele K, Phillips G. Preoperative magnetic resonance imaging cross-sectional area for the measurement of hamstring autograft diameter for reconstruction of the adolescent anterior cruciate ligament. Arthroscopy. 2008;24:1336–41. doi: 10.1016/j.arthro.2008.07.012. [DOI] [PubMed] [Google Scholar]
  • 3.Vivekanantha P, Grzela P, Wong S, Mansour F, Khalik HA, Johnson J, et al. Tendon cross-sectional area on magnetic resonance imaging and anthropometric characteristics can be used to predict insufficient four-strand hamstring autograft diameter in anterior cruciate ligament reconstruction: A systematic review. Knee Surg Sports Traumatol Arthrosc. 2024;32:1470–91. doi: 10.1002/ksa.12179. [DOI] [PubMed] [Google Scholar]
  • 4.Hodges CT, Shelton TJ, Bateni CP, Henrichon SS, Skaggs AW, Boutin RD, et al. The medial epicondyle of the distal femur is the optimal location for MRI measurement of semitendinosus and gracilis tendon cross-sectional area. Knee Surg Sports Traumatol Arthrosc. 2019;27:3498–504. doi: 10.1007/s00167-019-05421-6. [DOI] [PubMed] [Google Scholar]
  • 5.Ayres JM, Ose BM, Morey T, Brown E, Mar D, Henkelman E, et al. Preoperative magnetic resonance imaging measurements of hamstring tendons’ cross-sectional area may be used to predict the 5-Stranded graft diameter in anterior cruciate ligament reconstruction. Arthrosc Sports Med Rehabil. 2025;7:101001. doi: 10.1016/j.asmr.2024.101001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Kocazeybek E, Ersin M, Yılmaz M, Ekinci M. Preoperative estimation of hamstring autograft graft diameter in anterior cruciate ligament reconstruction with a new simple method using magnetic resonance imaging. Knee. 2025;53:193–9. doi: 10.1016/j.knee.2024.12.016. [DOI] [PubMed] [Google Scholar]
  • 7.Erquicia JI, Gelber PE, Doreste JL, Pelfort X, Abat F, Monllau JC. How to improve the prediction of quadrupled semitendinosus and gracilis autograft sizes with magnetic resonance imaging and ultrasonography. Am J Sports Med. 2013;41:1857–63. doi: 10.1177/0363546513479340. [DOI] [PubMed] [Google Scholar]
  • 8.Wotherspoon S, Giffin JR, Fowler P, Litchfield R, Neligan M, Willits K. Prediction of Anterior Cruciate Ligament Hamstring Autograft Size Using Preoperative Magnetic Resonance Imaging. In: 2009 Canadian Orthopaedic Association Annual Meeting. 2009:p. 241. [Google Scholar]
  • 9.Yan CC, Yang DL, Sun XY, Jiao Q. Correlation between preoperative MRI measurement of cross-sectional area of hamstring tendon and graft in anterior cruciate ligament reconstruction. Zhongguo Gu Shang. 2023;36:932–5. doi: 10.12200/j.issn.1003-0034.2023.10.005. [DOI] [PubMed] [Google Scholar]
  • 10.Shrestha P, Singh Thapa S, Mahara DP, Paudel S, Lamichhane A, Saud T. Prediction of hamstring autograft sizes for anterior cruciate ligament reconstruction using preoperative magnetic resonance imaging. J Nepal Health Res Counc. 2023;21:34–9. doi: 10.33314/jnhrc.v21i1.4339. [DOI] [PubMed] [Google Scholar]
  • 11.Wernecke G, Harris IA, Houang MT, Seeto BG, Chen DB, MacDessi SJ. Using magnetic resonance imaging to predict adequate graft diameters for autologous hamstring double-bundle anterior cruciate ligament reconstruction. Arthroscopy. 2011;27:1055–9. doi: 10.1016/j.arthro.2011.02.035. [DOI] [PubMed] [Google Scholar]
  • 12.Erquicia JI, Gelber PE, Doreste JL, Pelfort X, Abat F, Monllau JC. How to improve the prediction of the quadrupled hamstring tendon graft diameter in ACL reconstruction. Knee Surg Sports Traumatol Arthrosc. 2013;21:880–5. [Google Scholar]
  • 13.Yasumoto M, Ishibashi Y, Tsuda E, Sasaki K, Tazawa R, Yamamoto Y. Prediction of hamstring tendon autograft diameter using preoperative magnetic resonance imaging. Knee. 2021;29:222–7. [Google Scholar]

Articles from Annals of African Medicine are provided here courtesy of Wolters Kluwer -- Medknow Publications

RESOURCES