Therapeutic interventions for acute complete ruptures of the ulnar collateral ligament of the thumb: a systematic review

Introduction

Acute complete ruptures of the ulnar collateral ligament (UCL) of the thumb are common injuries, accounting for around 50 in 100,000 presentations to Accident and Emergency departments. There is controversy as how to manage complete ruptures of the UCL best, although there is a degree of consensus regarding the broader treatment algorithm and general agreement that ‘true’ Stener lesions should be managed operatively^1,2. The rate of the Stener lesion varies widely in the literature, perhaps reflecting the lack of reliability and accuracy of the various methods of diagnosis^3,4.

Patients should be assessed clinically to determine the degree of instability of the metacarpophalangeal joint (MCPJ) in both extension and 30° of flexion to test both proper and accessory collateral ligaments⁵. There is some evidence to suggest that the greater the instability the higher the chances are that a Stener lesion is present^6,7. While there is evidence to support both the use of ultrasound and MRI, the latter appears slightly superior in terms of sensitivity and specificity^8–10. A recent study by Stoop et al. investigated which factors predict the chances of surgery in UCL injuries¹¹. It was found that not only did patient characteristics influence the chances of surgery, but that the individual surgeon’s preference was also predictive.

Our aim was to perform a systematic review of the effectiveness of available interventions for acute complete rupture of the ulnar collateral ligament of the thumb in terms of patient-reported outcome measures and to assess the rates of adverse outcomes associated with these interventions.

Methods

The systematic review was developed in accordance with the PRISMA statement (Supplementary File 1 contains a completed PRISMA checklist), using methodology decribed in the Cochrane Handbook for Systematic Reviews of Interventions. The protocol was developed prospectively and peer reviewed locally before registration on the PROSPERO database (CRD42018087656).

Results

A total of 158 studies were identified by the search, after duplicates were removed. After screening by full-text, six studies were identified as eligible for inclusion (Figure 1). Of these, three were randomized controlled trials (RCTs), and three were retrospective comparative case series. The number of studies identified and excluded at each stage is detailed in Figure 1.

Figure 1. PRISMA flow diagram.

Study characteristics of the included trials including the interventions and comparators are provided in Table 1. Of the three randomised controlled trials, two assessed the outcomes of different rehabilitation regimes in patients who had been exclusively treated with surgery^13,14. The remaining RCT assessed the outcome in patients managed both surgically and non-surgically, who were randomised to treatment with either a plaster cast or a functional splint¹⁵. All three retrospective comparative case series compared different surgical techniques in patients exclusively managed surgically^16–18. Table 2 details the basic demographics of the intervention and comparator groups, as well as the details about the outcome data provided. The full details of all included studies and the forest plots are included within the supplementary material (Supplementary File 3–Supplementary File 15).

The study by Sollerman et al.¹⁵ compared a functional splint with plaster cast treatment in patients with complete UCL ruptures; patients were managed both surgically and non surgically. The authors reported no difference in MCPJ range of movement (ROM), grip strength and sick leave taken; however, the data provided were insufficient for any further analysis, such as a forest plot.

The RCT by Rocchi et al. compared the outcomes of operated patients treated with either a traditional standard thumb spica which immobilized the MCPJ or a new modified thumb spica which allowed early MCP motion¹⁴. At 12 months the new spica group had increased MCPJ ROM (standardized mean difference (SMD), −3.69; 95% confidence interval (CI), −2.46–−4.92, P<0.0001), a better Dreiser index (SMD, 1.65; 95%CI, 0.81–2.50; P=0.0001) and reduced pain VAS (SMD, 1.53; 95% CI, 0.70–2.35; P=0.0003). There was no statistically significant difference between groups in tip pinch strength at any time point. The RCT by Crowley et al. compared outcomes between patients treated with early active mobilization or plaster immobilization after being treated surgically with Mitek anchor repair¹³. The outcome data was not provided, meaning that any further analysis was not possible.

The retrospective comparative case series by Saetta et al. demonstrated a higher chance of an excellent functional result with suture repair versus steel wire, but this was not statistically significant (risk ratio, 1.19; 95% CI, 0.82–1.71); the other outcome data was incomplete and thus precluded further analysis. The retrospective case series by Lane demonstrated no statistically significant difference in the chances of a full versus partial recovery in ROM of the MCPJ, of a full versus partial recovery in strength and of a full versus partial functional recovery¹⁷. The study by Katolik et al. did not provide adequate data with which to conduct any further analysis¹⁶.

Risk of bias

All criteria were judged as low, high or unclear risk of bias. Overall, all studies were deemed to be at a high risk of bias, particularly in terms of blinding of outcome assessment and selecting reporting. Full risk of bias assessment is available in Figure 2 and Figure 3.

Figure 2. Risk of bias graph.

Figure 3. Risk of bias summary.

Discussion

The key finding of this systematic review is that is that no study exists comparing non-operative to surgical intervention in the treatment of complete ruptures of the UCL of the thumb. The only studies which have compared interventions are at high risk of bias, particularly in the areas of blinding of outcome assessment and selective outcome reporting. There is weak evidence to suggest that early mobilisation of the thumb MCPJ may be beneficial following surgical repair. There is weak evidence that the pull out suture fixation has a higher rate of adverse events when compared to bone anchor repair.

A systematic review by Samora et al. summarised the outcomes after both non-operative and operative treatment of complete UCL ruptures¹⁹. They found that the vast majority of the evidence base was low quality retrospective case series and that only a small minority of patients were treated non-operatively. It was also shown that there was no significant difference in outcome between repair of acute injury and reconstruction after chronic injury.

Landsman et al. demonstrated generally good results when managing complete ruptures with splintage with only 15% failing this regime non operative treatment⁴; notably, 30% of the patients in this series had displaced fractures and all patients had more than 30° laxity in 30° of MCPJ flexion. A case series reported by Pichora et al. also demonstrated generally satisfactory functional results with functional bracing, even in the 5 patients who were judged to have sustained true Stener lesions²⁰; notably, the three patients who failed functional bracing could not be predicted by the initial clinical tests. Case series purely relating to avulsion fractures of the UCL have shown contrasting results. For example Kuz et al.²¹ demonstrated satisfactory outcomes in all patients but a non union rate of 25%, this contrasts with the results of Dinowitz et al.²², which demonstrated poor functional results in patients treated non-operatively for minimally displaced fractures.

There is a widely varying rate of Stener lesions in the literature, it being as low as 12% in the series by Pichora et al.²⁰ and as high as 70% in other series²¹. The reasons underlying the variability in the rate of the Stener lesion are likely multiple and complex. One aspect of this conundrum appears to be the clear problems with the reliability and accuracy of the radiological diagnosis of the Stener lesion, particularly relating to MRI³. Although there are some high quality studies describing the reliability of ultrasound, there are no high quality studies relating to MRI³. Mahajan et al. demonstrated excellent agreement between radiologists in determining whether the UCL had completely ruptured; however, the presence or absence of a Stener lesion was not assessed radiologically⁶. Milner et al. have recently argued that any displacement of greater than 3 mm (grade 3 by their system) should be treated operatively, owing to the observed high chance that these patients will fail with non operative treatment⁹.

The recent study by Stoop et al. assessed 383 UCL injuries treated at three different hospitals in a singe American city¹¹. In total, 30% of cases were avulsion fractures and 11% of cases were investigated with an MRI scan. Certain patient characteristics were associated with a higher rate of operative intervention, for example greater age and more displaced fractures. However some factors which were unrelated to patient characteristics were predictive of operative intervention, such as having an MRI and being treated by certain surgeons. It was felt that because the preoperative diagnosis of a Stener lesion has limited reliability and accuracy, the rates of surgery may vary based on surgeon beliefs, preferences and values.They also stated that ”some surgeons believe some non-Stener injuries benefit from operative treatment”¹¹.

This review has demonstrated that all six studies of an intervention with a comparator in UCL treatment are at high risk of bias. The blinding of participants would clearly not be possible in a trial of surgery versus non operative treatment; however, it is a recurrent theme that outcomes were assessed by non-blinded assessors (often the treating surgeon), which significantly increases the chance that detection bias will influence patient outcomes. None of the RCTs published a trial protocol with a specified primary outcome, while only the study by Rocchi et al.¹⁴ used validated patient-reported outcome measures (Dreiser index and VAS). There was also a failure to adequately report all outcomes, with only one study reporting adequate data for all outcomes to allow further analysis. None of the three RCTs included a power calculation. While the retrospective nature of the comparative case series introduces several potential sources of bias which may have influenced these results.

Conclusions

There is no prospective randomised or observational evidence to support operative intervention compared to non-operative treatment for acute complete ruptures of the ulnar collateral ligament of the thumb. There is weak evidence to suggest that early mobilisation may be beneficial following surgical repair. Further research is necessary in order to better define which patients benefit from which specific interventions.

Data availability

All data underlying the results are available as part of the article and no additional source data are required.