Medium-Term Patient-Reported Outcomes after Surgical Management of Perilunate Injury: A Multiinstitutional Experience

Abstract

Background

There is a lack of consensus on the optimal treatment of perilunate injuries (PLIs). Open reduction with internal fixation with scapholunate (SL) ligament repair is recommended; however, procedures such as lunotriquetral (LT) ligament repair, carpal tunnel release (CTR), proximal row carpectomy, denervation, and arthrodesis may also be used.

Purpose

The purpose of this study was to assess patient-reported outcomes (PROs) after surgical management of PLIs.

Materials and Methods

A retrospective review of surgically treated PLIs at two institutions in a major metropolitan city from 2014 to 2020 was conducted. Demographics, radiographic data, and details of surgical management were reviewed. PROs were prospectively collected and analyzed.

Results

Eighty-seven patients with surgically managed PLIs were identified. The average time to surgery was 4 ± 7days. Twenty-six percent of patients presented with acute median neuropathy. Thirty-seven patients provided PROs for analysis. Patient-Rated Wrist Evaluation scores were 22 ± 12.9, 14.4 ± 11.3, and 37 ± 22 for pain, function, and total scores, respectively. Satisfaction was 7 ± 3 Sixteen percent of patients were unable to maintain employment due to their injury. The interval to return to work was 6.3 ± 7.7 months.

Conclusion

PLIs cause significant morbidity and affect patients' ability to return to work. While open reduction with SL repair is routinely performed to treat PLIs, additional procedures may be considered. Undergoing concurrent CTR in the absence of neurological symptoms showed no differences in outcomes.

Level of Evidence

Level III, retrospective cohort study.

Perilunate injuries (PLIs) are uncommon but devasting wrist injuries resulting from high-energy trauma. These injuries exist on a spectrum described first by Mayfield et al ¹ and later by Herzberg et al. ² Mayfield et al graded these injuries from stage I to IV, with a stage IV injury corresponding to the highest degree of ligamentous damage and carpal instability. The system described by Herzberg et al categorizes PLIs as stage I, where the lunate remains articulated with the radius, and stage II, where the lunate is dislocated from the radius. ^{1 2} PLIs can be further subclassified into greater arc and lesser arc injuries, with the former indicating the presence of a fracture with potential ligamentous damage and the latter indicating a purely ligamentous injury. ³ Most common in young men, PLIs are associated with high-energy mechanisms such as motor vehicle and motorcycle collisions, falls from height, and sports-related injuries. Consequently, many patients have concomitant injuries which are often life-threatening. ^{2 4 5 6} PLIs have been associated with poor functional outcomes, decreased quality of life, and inability to work. ⁷

Treatment of PLIs depends upon several factors including Mayfield/Herzberg classification, greater/lesser arc injury, other associated injuries, and presence of acute carpal tunnel syndrome. Though early attempts were made at nonsurgical and/or minimally invasive treatment of PLIs, the current consensus is that definitive management is surgical and includes reduction of the carpal bones (open reduction in cases that cannot be arthroscopically reduced), stabilization of the carpus, fixation of associated fractures, carpal ligament repair or reconstruction, and carpal tunnel release (CTR) in the presence of acute carpal tunnel syndrome. ^{5 8 9} There is significant variability in the literature with regard to the details of surgical management of PLIs including the surgical approach, ligament repair strategy and technique, and the optimal stabilization method. ¹⁰ Proximal row carpectomy (PRC) and wrist arthrodesis are also viable options in cases of primary treatment failure or severe carpal injury precluding typical surgical management. ^{2 11} In an effort to improve outcomes of PLIs, some surgeons perform additional prophylactic procedures such as wrist denervation and CTR in the absence of acute carpal tunnel syndrome, though there are little data to support this. ^{10 12} With the inherent heterogeneity of PLIs and little consensus on ideal surgical management, further study is needed to define the optimal treatment algorithm for these devasting wrist injuries.

The purpose of this study was to investigate the medium-term patient-reported outcomes (PROs) of surgically treated PLIs. A secondary objective of this study is to investigate the role of prophylactic procedures in the treatment of PLIs, including (1) CTR in the absence of acute median nerve symptoms and (2) primary wrist denervation. Our hypothesis was that satisfaction after surgical management would be acceptable and that prophylactic CTR and wrist denervation would demonstrate no improvement in PROs.

Materials and Methods

This research was approved by our Institutional Review Board prior to the initiation of the study (STUDY00003119). A review of all patients who underwent surgical management of PLIs from 2014 to 2020 at two separate institutions in a major U.S. metropolitan city was conducted. One institution is a safety-net level-I urban trauma center, while the other is a community-based private practice. All patients were treated by fellowship-trained hand surgeons. Demographics, injury characteristics, radiographs, and details of preoperative and operative management were recorded and analyzed. Radiographs were reviewed by the authors and assigned a classification based on both the Herzberg et al ² and Mayfield et al classification systems. ¹ All patients over the age of 18 years were included, and any patient who was identified to have been treated nonoperatively was excluded from the study. Any patient who did not provide PRO data was included in the demographics, treatment, and radiographic analysis but excluded from any analysis on outcomes.

Patients were contacted to collect Patient-Rated Wrist Evaluation (PRWE) scores, satisfaction and employment status, and details on revision operations performed at outside facilities. PRWE scores have previously been found to provide valid and consistent results for the measure of PROs following lunate injuries. ¹³ The PRWE is subdivided into two sections: pain (0–50) and function (0–50). Total PRWE scores are a sum of the pain and function scores ranging from 0 to 100. Higher scores indicate a worse outcome. Both pain and function subscores, as well as total PRWE scores, were collected and reported. Satisfaction was measured on a scale from 0 to 10, with 0 indicating complete dissatisfaction and 10 indicating complete satisfaction.

All analyses were performed via SPSS (Version 28.0.1). Independent samples t -test, chi-square, and Fisher's exact tests were performed where appropriate. Outcomes analysis of elective procedures was conducted with matched cohorts by age, sex, and Mayfield classification. Significance was set at 0.05.

Results

Demographics and Injury Patterns

Eighty-seven patients met inclusion criteria ( Table 1 ). Eighty-four percent of patients (73/87) were male, and the average age was 37 years (standard deviation [SD] 15). Thirty-six (41%) patients sustained lesser arc injuries, while 59% (51/87) were classified as greater arc. The most common associated wrist fractures involved the scaphoid (38%), radial styloid (22%), ulnar styloid (8%), and triquetrum (8%). Five percent (4/87) of injuries were missed at the initial presentation. Patients with lesser arc injuries were older compared with those with fractures (44.4 vs. 32.6 years, p  < 0.001). Acute median neuropathy was diagnosed in 26% (23/87) of patients at presentation. The most common mechanism of injury involved a motor vehicle (car, motorcycle, all-terrain vehicle; Table 1 ). While 56% (49/87) of patients presented with isolated wrist injuries, 44% (38/87) of patients had concomitant orthopaedic or multisystem injuries. Admission to the intensive care unit was required in 21% (18/87) of the patients. The most common associated injuries were orthopaedic (40%), face/ear nose throat (13%), thoracic (13%), and spine (10%). Other orthopaedic injuries present included fractures of the radius/ulna (16%), tibia/fibula (11%), pelvis (7%), humerus (6%), femur (6%), and scapula (3%).

Table 1. Demographics and injury data for perilunate injuries.

Age (years)	37 (15)
Sex	N  = 87
M	73 (84%)
F	14 (16%)
Dominant side	27 (52%)
Race	N  = 87
African American	53 (61%)
Caucasian	22 (25%)
Hispanic	8 (9%)
Asian	1 (1%)
Other	3 (3%)
Mechanism of injury	N  = 87
Motor vehicle	48 (55%)
Fall	24 (28%)
Sporting injury	9 (10%)
Assault	4 (5%)
Other	2 (2%)
Reduction attempt	N  = 76
Success	32 (37%)
Fail	44 (51%)
Median nerve symptoms	N  = 87
None	53 (61%)
Mild	11 (13%)
Severe	23 (26%)
Time to surgery (days)	4 (7)

Note: Data presented as mean (standard deviation) or count (%).

Sixty-seven patients were assigned a classification based on Mayfield et al ¹ and Herzberg et al ² ( Table 2 ). The radiographs of five patients were determined to be unclassifiable due to the severity of the injury and anatomical disruption. The 15 remaining patients did not have preoperative radiographs available to review, the vast majority of which were referred from an outside facility and treated at the community-based private practice. Mayfield III and IV injuries were seen in 84% (56/67) and 16% (11/67), respectively. As per Herzberg, 70% (47/67) and 24% (16/67) of injuries were classified as dorsal 2a or dorsal 2b, respectively, and 6% (4/67) had a volar dislocation of the carpus relative to the lunate.

Table 2. Radiographic classification of perilunate injuries.

Herzberg classification	N  = 67
Dorsal 2a	47 (70%)
Dorsal 2b	16 (24%)
Volar (1, 2a, 2b)	4 (6%)
Mayfield classification	N  = 67
III	56 (84%)
IV	11 (16%)

Preoperative Reduction and Surgical Treatment

Preoperative reduction of the PLI was attempted in 87% (76/87) of the patients ( Table 1 ). A successful reduction was defined as the maintenance of near anatomic alignment of the carpus with restoration of lunocapitate congruency. No reduction was attempted in 13% (11/87) of the patients due to a missed diagnosis or due to the severity of injuries precluding a successful reduction. Only 42% (32/76) were able to be successfully reduced prior to operative treatment. A failed reduction occurred in 58% (44/76) of the patients. Fifty-two percent of patients underwent surgical operation within the first 24 hours after presenting. The average time from injury to surgery was 4 days. There was no difference in time to surgery between a successful reduction and nonsuccessful reduction (4.9 vs. 3.7 days, p  = 0.545).

Table 3 summarizes the surgical techniques employed. Most injuries were treated surgically with either a dorsal-only approach (37/87, 42%) or combined dorsal/volar approach (47/87, 54%). Closed reduction and percutaneous pinning alone was performed in 2% (2/87). A volar approach for CTR and CRPP of the lunate was used for one patient. All patients underwent a scapholunate (SL) repair when indicated, while the choice to repair the lunotriquetral (LT) ligament was based on surgeon preference. Forty-four percent of greater arc injuries required a repair of the SL ligament, while 29% of these patients underwent a repair of the LT ligament. Ninety-one percent of patients with lesser arc injuries required a repair of the SL ligament, while 32% of these patients had a repair of the LT ligament. Three percent (3/87) were treated with an acute PRC, and 1% (1/87) of patients were treated with a four-corner-arthrodesis as primary treatment. Kirschner wires (K-wires) were used in 89% (78/87) of the patients to provide stability to the carpus or to protect a ligamentous repair. Eleven percent of repairs without K-wire fixation were isolated transscaphoid fracture dislocations or patients treated with PRC or four-corner-fusion without concomitant radial or ulnar styloid fracture. When a scaphoid repair was indicated, headless compression screws were utilized in addition to any ligamentous repair or additional fracture fixation.

Table 3. Surgical characteristics.

	Number (Percent)
Approach
Dorsal	37 (42)
Volar + Dorsal	47 (54)
Ligament repair
SL repair	55 (63)
LT repair	26 (30)
Associated procedures
CTR	45 (52)
Denervation	23 (26)
Acute salvage procedures
PRC	3 (3)
4CA	1 (1)

Abbreviations: 4CA, 4-corner arthrodesis; CTR, carpal tunnel release; LT, lunotriquetral; PRC, proximal row carpectomy; SL, scapholunate.

Outcomes

Thirty-seven patients participated in the postoperative PRO data. The average age was 41 (SD, 15.8) years, and the group had 89% males. Fifty-seven percent (21/37) were greater arc injuries versus 43% (16/37) lesser arc injuries. The SL was repaired in 65% (24/37), and the LT was repaired in 35% (13/37). Thirty-one patients had preoperative X-ray for review. Fifty-four percent (20/31) were categorized as Mayfield III and 22% were Mayfield IV. By Herzberg classification, 46% and 27% were dorsal 2a, respectively. One patient had a volar dislocation. Three patients had unclassifiable injuries. The average time from injury date to survey collection was 32 months and ranged from 12 to 63 months. Patients had an average satisfaction of 6.6/10, and the average PRWE score was 36.0 at the final follow-up ( Table 4 ). PRWE pain and function subscores were 21.6 and 14.3, respectively. After injury, 76% (28/37) were able to return to work and perform the same job duties as before. Three were able to work but had to change jobs due to his wrist injury. Six patients (16%) were unable to return to any kind of work due to their wrist. The average interval between injury and return was 6.35 months in 31 patients.

Table 4. Outcomes after perilunate injuries.

	PLD	PLFD	Total	p -Value
PRWE
Total	44.9 (24.8)	33.6 (22.7)	36.0 (22.2)	0.284
Pain	25.5 (14.4)	21.2 (11.7)	21.6 (13.0)	0.449
Function	19.1 (11.9)	12.2 (12.2)	14.3 (11.1)	0.206
Satisfaction	6.8 (3.9)	6.7 (3.2)	6.6 (3.1)	0.956
Return to work (months)	3.35 (2.8)	4.7 (3.8)	6.35 (7.7)	0.560
Treatment failure	1 (6%)	1 (5%)	2 (5%)

Abbreviations: PLD, perilunate dislocation; PLFD, perilunate fracture dislocation; PRWE, Patient-Rated Wrist Evaluation.

Note: Data presented as mean (standard deviation) or count (%).

p -Value is calculated between PLD and PLFD groups.

Lunotriquetral Ligament Repair

The LT ligament was repaired in 30% (26/87) of patients. No differences were identified in the PRWE scores or satisfaction of patients with LT repair compared with those without (pain, 25.3 vs. 20.1, p  = 0.411; function, 11.3 vs. 12.0, p  = 0.872; total, 36.8 vs. 32.3, p  = 0.668; Table 4 ).

Carpal Tunnel Release

The carpal tunnel was decompressed in 52% (45/87) of the patients. Twenty-three patients presented with significant median nerve symptoms indicative of acute carpal tunnel syndrome and were treated with urgent surgical intervention. Six patients underwent a CTR as part of the volar approach to a capsular or ligamentous repair. Sixteen additional patients without indications for acute carpal tunnel decompression or soft tissue repair, including 36% (4/11) of patients with mild median nerve symptoms, underwent CTR during the index procedure as a prophylactic. CTR in these cases was considered elective. No significant differences were seen between PRWE and satisfaction scores in patients with elective CTR compared with no CTR (pain, 26.5 vs. 19.9, p  = 0.265; function, 17.9 vs. 11.0, p  = 0.185; total, 44.6 vs. 31.1, p  = 0.259). No significant differences in satisfaction were seen between groups (6.13 vs. 6.63, p  = 0.954; Table 5 ).

Table 5. Outcomes after elective procedures in perilunate management.

	+ LT Repair	−LT Repair	p -Value	+ CTR a	−CTR a	p -Value	+Neurectomy b	−Neurectomy	p -Value
PRWE
Pain	25.3 (12.4)	20.1 (13.3)	0.411	26.2 (15.7)	19.9 (11.9)	0.265	20.7 (22.8)	22.8 (12.0)	0.353
Function	11.3 (7.2)	12.0 (11.0)	0.872	17.9 (13.9)	11.0 (10.5)	0.185	11.9 (10.1)	15.0 (11.6)	0.266
Total	36.8 (19.4)	32.3 (23.6)	0.668	44.6 (28.5)	31.1 (20.6)	0.259	32.9 (26.7)	38.0 (21.7)	0.298

Abbreviations: CTR, carpal tunnel release; LT, lunotriquetral; PRWE, Patient-Rated Wrist Evaluation.

^aExcludes those presenting with acute carpal tunnel or CTR for a volar approach to the wrist.

^bAnterior interosseous nerve/posterior interosseous nerve neurectomy.

Elective Wrist Denervation

Twenty-three (26%) patients underwent wrist denervation via an anterior interosseous and/or posterior interosseous neurectomy. There was no significant difference in PRWE scores between those who received a neurectomy and those who did not (pain 20.7 vs. 22.8, p  = 0.353; function 11.9 vs. 15.0, p  = 0.266; total 32.9 vs. 38.0, p  = 0.298; Table 5 ).

Discussion

This study presents the demographics, acute management, and medium-term PROs of PLIs from 2014 to 2020 at two institutions in a large metropolitan region. Nonoperative management of PLIs has been associated with poor outcomes; however, there is no consensus in the literature as to the ideal operative management of these injuries. ^{14 15} Open reduction, internal fixation (ORIF) with SL ligament repair, when indicated, has been described as the “gold standard” for PLIs. ^{6 16 17 18} However, there have been multiple studies which challenge this dogma. Forli et al demonstrated anatomic SL gaps in 35 of 39 patients treated with K-wire stabilization and only partial or no SL repair. ¹⁹ An isolated volar approach with no ligamentous reconstruction has also been associated with satisfactory outcomes without additional complications or revision surgeries. ⁸ This lack of consensus within the literature adds to the complexity of PLIs. While the SL ligament was repaired in all indicated cases in this series, an LT ligament repair was not considered standard treatment. Repair of the LT ligament is theoretically protective against the development of volar intercalated segment instability (VISI); however, several studies have shown no increased incidence of VISI in patients who do not undergo LT repair if an anatomic reduction is achieved, and the joint is stabilized with K-wires. ^{12 19 20} LT ligament repair did not have a significant effect on PROs in this series, though the authors recognize the limited utility of this finding given the lack of objective outcome measures and relatively short-term follow-up. The benefit of LT ligament repair in PLIs remains an unanswered question and warrants further prospective investigation.

Even with successful treatment, patients may have noticeable functional deficits and persistent pain as a result of the PLI. ^{10 21 22 23} Prognostic factors for worse outcomes include delays in management or missed diagnosis, persistent carpal malalignment, open injuries, and osteochondral fractures of the carpus. ¹⁰ Significant arthritic degenerative radiographic changes have been reported in 70% of wrists at 15-year postinjury; however, radiographic degeneration has not correlated with worsened clinical outcomes. ^{6 19 24} Patients have demonstrated between 13 and 30% decrease in functional grip strength after surgical treatment. ^{10 22} While 88% of patients in this series were able to return to work, only 76% were able to return to the same job or the same functional duties due to residual limitations from their injury. After definitive treatment, most patients were unable to work for months with an average return to work of more than 6 months in this study. These results are similar to existing literature with the return to work in physically demanding work within 6 to 12 months with focused rehabilitation to regain adequate function. ²⁵

Few studies have differentiated the outcomes of PLD and PLFD injuries; however, pure ligamentous injuries are suggested to have worse outcomes due to the loss of stable carpal alignment from soft tissue disruption. ^{6 7} Laporte et al published a series of 17 PLIs showing significantly worse clinical outcomes (difference of 46 in total PRWE scores) and worse radiological outcomes in pure dislocations than fracture dislocations. Bagheri et al further described a similar trend, albeit with less substantial differences in surgically treated PLIs—mayo wrist score 88 versus 85, flexion 55 versus 50 degrees, and extension 54 versus 52 degrees. While no differences met the threshold for significance in this study, there was a nonsignificant trend toward worse functional and pain outcomes in patients with pure ligamentous injuries. Despite the worse trends in clinical measure for PLD, the near identical satisfaction between PLD and PLFD injuries may suggest that any differences in the patients' injury or their outcome are not functionally noticeable for the patients in their day-to-day lives. Regardless of the classification or the severity of the injury and despite the fact that most patients in this study were able to return to work, patients should be counseled that poor outcomes including persistent pain and loss of motion can be expected postoperatively or develop in the coming years. The posttraumatic degeneration that may occur after these injuries presents a significant burden on patients' lives which should be counseled to patients during treatment to set expectations.

Salvage surgeries such as PRC and wrist arthrodesis have been traditionally reserved for patients who develop late complications after PLIs; however, there seems to be a minority of patients with particularly severe injuries who benefit from these operations as a primary treatment option. ^{24 26} Huish et al ¹⁶ recommended PRC in the acute setting where anatomic fixation would be difficult or unlikely and Bain et al ⁴ also reported on primary salvage of PLI injuries with either PRC or isolated lunate excision for injuries that presented outside of the acute window; however, consistent follow-up data on subsequent outcomes were lacking. In a series of 21 cases, Muller et al ²⁷ compared the medium-term outcomes between ORIF and PRC for the primary treatment of PLIs. While those treated with ORIF were more likely to have been treated acutely from the time of injury (1.2 vs. 18.7 days), PRC-treated PLIs had comparable medium-term results in their grip strength, range of motion, Quick-DASH, and PRWE scores. Five percent (4/87) of the patients in our series underwent a salvage operation at the index procedure. These patients presented with multiple highly comminuted carpal bone fractures and unclassifiable radiographs. Despite the significant disruption of the carpus, they all reported excellent pain and functional outcomes at 36 months. While the sample sizes were not amenable to statistical analysis, the patients treated with arthrodesis or PRC had 10/10 satisfaction and PRWE scores of 5, 5, and 10 for pain, function, and total, respectively. While we do not recommend the routine use of salvage procedures in the primary treatment of PLIs, when anatomic reduction is unlikely due to fracture comminution or soft tissue injury, a primary salvage operation may be a reasonable procedure.

The benefit of additional procedures during the primary treatment of PLIs remains unclear. Some surgeons prophylactically release the carpal tunnel in patients with PLIs regardless of median nerve symptoms given the inherent severity of carpal injury and perceived risk to the median nerve. To the best of our knowledge, no studies have investigated the benefits of “elective” CTR in PLI. In this study, there was no significant difference in outcomes between patients who underwent elective CTR and those who did not, but there was a notable trend toward better outcomes in the group who did not undergo CTR. The significance of this trend remains unclear. There may have been a selection bias toward CTR in patients with a greater perceived severity of injury, which would potentially explain the inferior outcomes in this group. It is also possible that CTR in the absence of acute median neuropathy simply causes additional trauma and subsequent scarring without clear benefits (it is worth noting that there were no instances of postoperative carpal tunnel syndrome requiring intervention in patients who did not undergo CTR). A combination of the volar–dorsal approach in the treatment of PLIs has been suggested to potentially contribute to further surgical trauma, swelling, and carpal devascularization. ²⁶ Further research is needed to better elucidate the impact of “elective” CTR in patients with PLIs.

Wrist denervation remains a surgical solution for patients with chronic wrist pain as it maintains function and mobility while denervating the primary fibers that innervate the wrist capsule. ²⁸ Denervation has been shown to provide pain relief as well as improve grip strength and DASH scores. ^{29 30} Denervation of the wrist joint has been utilized as an adjunctive treatment for traumatic and degenerative conditions of the wrist. The efficacy of denervation has been shown to decrease with time, but the quality of wrist function remains satisfactory in long-term outcomes. ^{28 31} Primary wrist denervation as an adjunctive procedure in the acute treatment of PLIs has not been previously reported in the literature. Given the relatively low morbidity of wrist denervation and that radiographic posttraumatic arthritis occurs in at least 70% percent of the patients following PLI, ^{6 19 24} it may be beneficial to prophylactically denervate the wrist in these patients. In the current study, PROs trended positively in patients who underwent prophylactic wrist denervation, but this did not achieve statistical significance. Further investigation is needed to determine the role of wrist denervation in the primary treatment of PLIs, but the authors believe that it is an adjunct worth considering given the high rate of posttraumatic arthritis in these patients. However, it should be noted that the observed trend toward improvement in this study falls below the mean clinically important difference for the PRWE which further calls into question the utility of primary denervation.

While this study presents a large longitudinal sample of patients treated from two institutions, there are several limitations that should be noted. Apart from survey collection, this was a retrospective chart review and subject to the inherent limitations of this study type. Incomplete survey participation may not completely reflect the entirety of the cohort. Objective measures including functional assessments and radiologic data could not be analyzed in this study secondary to the low rate of in-person follow-up data within the cohort.

Conclusion

Despite recent advances in the management of PLIs, there remains no consensus on the optimal surgical approach. This study provides further evidence of the inherent variability of injury patterns and surgical techniques in PLI. Most patients are able to return to work and report moderate satisfaction with wrist function. Lunotriquetral ligament repair, prophylactic CTR, and primary wrist denervation did not improve PROs in this cohort. Additional research is necessary to evaluate standardized approaches to PLI management and incorporate new surgical techniques to improve outcomes.

Funding Statement

Funding None.