Perilunate Dislocation Reduction Technique and Results

Abstract

Background:

Perilunate dislocations are severe upper extremity dislocations to the wrist involving ligamentous injury. Previous literature cites closed reduction of perilunate injuries at 25%. Herein, the technique and results of closed reductions for perilunate dislocations are described.

Methods:

Chart reviews were conducted of perilunate injuries treated from 2017 to 2022. Imaging was evaluated to categorize by Herzberg stage. Reduction was performed in the emergency department (ED) and involved traction, relaxation, and manipulation. Reduction attempts and success rates were recorded as were presence and resolution of median nerve symptoms.

Results:

Fifty-six perilunate injuries were included with an average age of 36 years. Forty-five (80%) underwent attempted closed reduction, of which 37 (82%) were successful. Twenty-seven (48%) were Herzberg I, 18 (32%) were Herzberg IIA, and 11 (20%) were Herzberg IIB. Two Herzberg IIB injuries were mal-reduced, with the lunate returned to the lunate fossa inverted. Scaphoid fractures were more likely in Herzberg I injuries and did not influence the success of reduction. Thirty (54%) patients presented with median nerve symptoms, of which 25 underwent attempted closed reduction in ED. Perilunate reduction resolved nerve symptoms in 92% of cases. Nerve symptoms were not more common in Herzberg II perilunate injuries. Successful closed reduction of perilunate injuries was not associated with a specific Herzberg stage.

Conclusion:

Closed perilunate reductions in the ED can have a high success rate >80%. Herzberg II perilunate injuries have >70% (76%) success rate in closed reduction. Rates of median nerve symptoms were high and resolved in most cases (92%) after reduction. Closed reduction of perilunate injuries can safely be attempted in the ED and resolves nerve compression prior to surgery.

Introduction

Perilunate dislocations are rare high-energy injuries resulting in disruption of the carpal anatomy. The characteristic feature is dorsal dislocation of the capitate from the concave surface of the lunate. More rare manifestations involve complete volar dislocation of the lunate from the lunate fossa of the radius. Perilunate injuries typically leads to sequential instability of the scapholunate, lunocapitate, and lunotriquetral articulations and can be divided into lesser arc injury (pure ligamentous disruption) and greater arc (combination of fracture(s) and ligamentous injury). The Herzberg classification of perilunate injuries is radiographic. Herzberg stage I refers to dorsal dislocation of the capitate from the lunate (Mayfield I-III). Herzberg stage II classification represents Mayfield IV injuries and is further subdivided into stage IIA, with <90° of lunate volar rotation, and stage IIB, with >90° of lunate volar rotation ¹ (Supplemental Figure 1).

Due to the relatively limited space in the wrist, these injuries can lead to acute compression of the median nerve in about 25% of cases.^{2 -5} Initial treatment entails immediate reduction to relieve tension on soft tissues and nerves and minimize swelling. Operative management is later performed with early surgical stabilization.^6,7 However, perilunate dislocations can be difficult to reduce, especially variants involving complete lunate dislocation (Herzberg stage II). Historically, reports of successful closed reduction rates are around 25%, ² with no recent literature update on the subject. The technique was originally described by Tavernier; however, the original description is not in the online literature and no detailed description is published. Herein, the technique and results of closed reductions for complex perilunate dislocations, including Herzberg type IIB complete lunate dislocations from the lunate fossa, are described.

Method for Reduction

Muscle relaxation is an essential component to the success of a closed reduction for a perilunate dislocation, particularly for Herzberg IIB injuries. To accomplish this, a combination of axial traction and conscious sedation is employed. The degree of medication necessary for obtaining adequate relaxation versus muscle fatigue is dependent on the patient (both their structure and stability for receiving medications) and physician’s expertise.

A combination of propofol/ketamine is administered by a physician trained in the use of conscious sedation. The choice of conscious sedation medication can also influence the objective of achieving muscle relaxation and thus should be discussed by the sedation and reduction team. Appropriate sedation is achieved when the patient does not resist the reduction procedure.

The affected extremity is first suspended in traction for at least 10 minutes, the patient is then sedated, and the reduction is performed. Sedation is used after applying traction to limit the time and use of sedation, thus minimizing its associated risks. This further allows repositioning of the patient if needed from potentially the seated position for traction and then supine for sedation.

Technique

Positioning

The patient can be supine, reclined at 45°, or seated upright in a comfortable position. Patient positioning will be dependent on the patient’s stability and the physician’s choice of medications. All jewelry should be removed from the patient’s affected limb, including rings and bracelets. Intravenous access in or distal to the antecubital fossa should be removed from the affected limb prior to reduction.

The patient’s shoulder should be flexed if in a seated position (if supine, shoulder flexion is not needed) and abducted 90° with the elbow bent at 90°. Their forearm should therefore be perpendicular to the ground, with fingers pointed vertically to the ceiling (Figure 1).

Figure 1.

Setup for adequate traction to properly reduce a perilunate dislocation.

Traction

There must be adequate axial traction and muscle fatigue at the wrist. This is achieved by first hanging the patient in traction prior to sedation. If a finger trap appliance is not available, one can be created with Kerlix (Byram Healthcare) wrap. The point of origin for the traction (ie, intravenous pole) is positioned directly above the patient’s elbow. It is important to align this correctly to avoid any rotational forces around the wrist. A loop of gauze wrap is created around an intravenous pole or overhead lamp with the knot at the top. The loop is pulled between the patient’s index and middle finger (Figure 2a). Then, this loop is folded back up over itself, wrapping around the 2 fingers (Figure 2b). This will create a self-tightening loop around the index and middle fingers individually (Figure 2c). As the weight of the patient’s arm is hung, the finger trap will tighten, applying axial traction (Figures 2d and 3). This traction alone is usually insufficient, and therefore, weight must be added. The patient is left in traction for 10 to 15 minutes to fatigue musculature prior to sedation and reduction.

Figure 2.

Technique for makeshift traction.

Note. (a) A loop of Kerlix is made and brought between the patient’s index and long fingers. (b) The loop is brought dorsally over fingers. (c) The Kerlix is tightened leading to (d) traction on the fingers. (e) The anterior humerus is well padded and (f) 4.5kg is used to fatigue muscles around the wrist.

Figure 3.

Line art depicting technique for makeshift traction.

Reduction

In some cases, the traction alone will spontaneously reduce the dislocation. The wrist should therefore be palpated again and examined for continued dislocation prior to attempting the reduction. If available, portable fluoroscopy can be used to confirm reduction or continued dislocation. Lateral images should be obtained. Often, formal reduction techniques (described below) are necessary, even after traction of the affected limb.

Herzberg stage I: In these injuries, it may be necessary to prevent volar dislocation of the lunate with reduction. This is achieved by gentle volar-to-dorsal pressure on the lunate to keep in the fossa. The lunate is located just proximal to the volar wrist crease and gentle volar-to-dorsal pressure is used to hold in the fossa. After the lunate is located and stabilized, the wrist is reduced with hyper-extension, axial traction, and subsequent flexion.

Herzberg IIA&B—lunate dislocated from the fossa: The dislocated lunate is palpated—typically at the proximal to the level of the wrist crease (Figure 4a). The most common bony block to reduction is the dorsal lip of the lunate hinging on the proximal pole of the capitate. To overcome this, the wrist is hyperextended to remove the capitate from the lunate fossa while dorsal pressure is applied to the volar aspect of the lunate (Figure 4b and c). This allows the dorsal lunate reduce underneath the capitate. Then, in one continuous motion, additional axial traction is applied to bring the capitate distally while the wrist is flexed. During this maneuver, gentle pressure is maintained on the lunate, which should reduce back into the fossa with the capitate reduced into the concave distal lunate (Figures 4d, e and 5). The splint is then applied with the wrist in neutral to slight flexion and the reduction is confirmed with 3-view wrist films (Figure 6a-e). The choice of splint (eg, sugar tong, volar slab) is at the physician’s discretion as the patient’s injuries and hospital regulations might dictate options. Regardless of the specific type, the splint should ensure immobilization of the carpus. The wrist must not be over-flexed (“cotton-picker” position) to avoid irritation of the median nerve. If hyperflexion is required to maintain the reduction, open reduction and fixation should be performed. If the reduction is unsuccessful, the authors recommend only one additional attempt to avoid exacerbation of soft tissue swelling.

Figure 4.

Reduction attempt.

Note. (a) The lunate is palpated in the volar wrist. (b and c) With gentle pressure on the lunate, the wrist is hyperextended. (d and e) Gentle pressure is maintained on the volar lunate, and with continued traction, the wrist is brought back into flexion.

Figure 5.

Line art depicting technique for reduction attempt.

Figure 6.

Splint technique.

Note. A sugar tong splint is used at our institution; however, a volar, dorsal, volar/dorsal splint may also be used. (a and b) The arm is maintained in traction and padded with Webril. (c) The splint is placed. (d and e) ACE wrap is used to secure the splint to the arm. (f and g) The Webril is cut and trimmed. (h) Ensure the traction is maintained while the mold settles.

Postreduction

Immobilization is achieved with splint placement with a gentle mold with the wrist in slight flexion. After the splint has hardened, traction is released by cutting the Kerlix or removing the finger traps and the affected extremity is left elevated (Figure 6f-h). For patient comfort, avoid allowing the affected extremity to be left in traction longer than 20 minutes. A repeat neurovascular examination is performed after traction is released and patient awake to confirm that there is resolution of neurovascular compromise. If prereduction neurovascular symptoms persist after reduction, an acute carpal tunnel release was performed. Until surgery, repeat neurological examinations are performed to monitor for development of median nerve symptoms. Signs and symptoms requiring urgent carpal tunnel release include: significant increasing pain within the median nerve distribution, loss of 2-point discrimination within this distribution, and paralysis of opposition. Regardless of reduction success and neurovascular examination, all patients underwent subsequent surgical fixation of their injury.

Outcomes

Methods

A retrospective chart review of all perilunate injuries treated by the orthopedics service in an urban level 1 trauma center from 2017 to 2022 was performed. Prereduction radiographs and postreduction radiographs and computed tomography were assessed for reduction by a hand fellowship–trained board-certified orthopedic surgeon. Reduction was deemed successful if the lunate had completely returned to the fossa and capitate to the distal lunate.

Median nerve symptoms were assessed by the orthopedic resident in the emergency department (ED) and were deemed to be present if there were paresthesias in the median nerve distribution, decreased sensation to light touch, severe pain in median nerve distribution, or paralysis of thumb opposition. Postoperative median nerve symptoms were examined similarly.

Results

There were 56 patients who underwent assessment by an orthopedic resident in the ED between 2017 and 2022. Forty-five patients had attempted closed reduction in the ED, with an overall success rate of 82%. Twenty-seven (48%) were Herzberg stage 1, 18 (32%) were Herzberg stage IIA, and 11 (20%) were Herzberg stage IIB. Of the Herzberg stage I injuries, 20 underwent an attempted closed reduction, of which 18 (90%) were successful. Of the Herzberg stage IIA cases, 16 underwent an attempted closed reduction, of which 12 (75%) were successful. Of the Herzberg stage IIB, 9 underwent an attempted closed reduction, of which 7 (78%) were successful. Two Herzberg stage IIB injuries were mal-reduced, with the lunate reduced in the fossa inverted. Successful closed reduction of perilunate injuries was not associated with a specific Herzberg stage (P = .222; Table 1).

Table 1.

Perilunate Reduction Success and Nerve Symptoms by Herzberg Category.

Perilunate Injury Characteristics	All injuries	Herzberg I	Herzberg II
			A	B
Number	56	27 (48%)	18 (32%)	11 (20%)
Attempted reduction	45 (80%)	20 (74%)	16 (89%)	9 (82%)
Successful reduction	37 (82%)	18 (90%)	12 (75%)	7 (78%)
Scaphoid fracture	22 (39%)	15 (56%)	4 (22%)	3 (27%)
Median nerve symptoms	30 (54%)	14 (52%)	11 (61%)	5 (45%)
Prereduction attempt nerve symptoms	25 (56%)	10 (50%)	10 (63%)	5 (56%)
Postreduction attempt nerve symptoms	2 (8%)	1 (5%)	0 (0%)	1 (11%)

Twenty-two injuries (39%) involved a scaphoid fracture. Scaphoid fractures were more likely in Herzberg stage I injuries than II (P = .016). All patients had median nerve examination documented prereduction and postreduction. Thirty (54%) patients presented with prereduction median nerve symptoms. Twenty-five of these patients underwent an attempted closed reduction in the ED. Twenty-three patients (92%) had resolution of median nerve symptoms postreduction. Median nerve symptoms were not found to be more common in Herzberg II perilunate injuries (P = .803).

Of the 45 closed reduction attempts, 44 had data listing the sedation medications. Of the 43 patients who received sedation, 23 (53%) received ketamine, 21 (49%) received fentanyl, 16 (37%) received propofol, and 4 (9%) received midazolam. Nine (21%) patients received a hematoma block. In terms of sedation, ketamine alone had 100% (7/7) success rate, fentanyl alone had 75% (6/8) success rate, propofol alone had 50% (1/2) success rate, and no patient had midazolam alone. Twenty-two patients received 1 sedative, 15 patients received 2 sedatives, and 4 patients received 3 sedatives. The number of sedatives did not affect the reduction outcome (P = .183). The individual sedatives did not affect reduction outcome: ketamine (P = .854), fentanyl (P = .346), propofol (P = .485), midazolam (P = .503).

Discussion

While the agreed-upon gold standard definitive treatment for perilunate injuries is surgical, there remains contention concerning whether attempting closed reduction is warranted. Even more so, whether to pursue emergent open reduction versus trial of closed reduction for Herzberg stage II injuries or those with median nerve symptoms is of particular discord. Better access to operating theaters over time may drive the treatment away from pre-op reduction attempts and toward immediate definitive management. Inconsistencies in successful reduction attempts may have further emphasized this transition.

The above-described reduction technique is a simple and effective method for alleviating median nerve symptoms, allowing for conversion from emergent to urgent surgical management. Re-emphasizing the role of this reduction technique is vital to trauma orthopedic surgeons. The successful reduction rates presented above challenge the notion that Herzberg stage IIB perilunate dislocations are irreducible or should not be reduced. Urgent reduction in the ED should be attempted with high success rates and resolution of median nerve symptoms. This allows more time for surgical planning, staffing with the most appropriate surgical team, acquisition of necessary equipment, and methodical scheduling. Furthermore, for patients in remote locations without immediate access to surgical intervention, reduction prior to long-distance transfer or awaiting follow-up may improve long-term outcomes.

In this series, there were 2 out of 9 instances of reduction of Herzberg IIB injuries resulting in the lunate inverted in the fossa. While Herzberg recommended that no IIB be closed reduced due to possibility of vascular-ligamentous injury during reduction attempt, this complication occurred in patients where the lunate was grossly removed from the fossa (in the volar mid forearm or dorsal hand). ⁶ Based off the above case series data, reduction is recommended in these cases only if there is neurovascular compromise with a possible delay to operating room, as there is the possibility of potential damage to the remaining articular cartilage or ligamentous structures which may preclude other wrist motion sparing treatment. Regardless of closed reduction timing—if any—surgical intervention is still recommended for definitive treatment and to ensure best possible functional outcomes for patients.

Literature describing closed reduction for perilunate injuries often limits its use to exclude Herzberg stage IIB injuries, suggesting only open reduction for these cases. This follows the notion that these are irreducible injuries. Regardless, the reduction technique used in these cases for this series is the same as that used for Herzberg stage I-IIA. Of note, the technique described above does not include local anesthetic (eg, hematoma block), unlike some other reports. Perhaps, the avoidance of adding additional fluid to the area of injury avoids increasing pressure in the area, thus making the reduction easier and improving success rates. This divergence in reduction technique may explain the difference between successful and failed closed reductions for such complex perilunate injuries. Regardless of local block use, it is universally agreed upon that adequate relaxation is necessary to be successful with perilunate reductions. Most commonly, similar to the technique presented above, conscious sedation is employed in closed reduction of perilunate injuries. The choice of pharmacology for attaining adequate relaxation affecting the success of the reduction remains to be explored, as the medications are often dependent on a myriad of factors including the patient’s presentation and the clinician’s comfort with certain combinations.

Reports avoiding closed reductions of Herzberg stage IIB injuries primarily cite inability to maintain the successful reduction if it is obtained, prompting the recommendation to avoid the procedure altogether and rush to open definitive treatment instead. ⁸ However, the data presented above had successful maintenance of the reductions in these cases. The splints applied relied on creating a mold around the lunate, countering its tendency to displace, maintaining the reduction. It is important to note that the entire wrist position must be considered to avoid applying pressure on the carpal tunnel, enhancing risk to the median nerve. The combination of the above points highlights how obtaining and maintaining closed reductions for these injuries is in part skill-dependent. Clinicians with less access to immediate operating theaters for perilunate injuries are more likely to attempt such reductions and place these splints, encouraging their likelihood of being successful with their increase in repetition.

Literature reports a variety of median nerve symptom rates in perilunate injuries. Several studies have suggested acute median nerve symptoms ranging from 23% to 45%, with the data above consistent with the higher incidence. ⁸ Delayed treatment may increase the prevalence of nerve symptoms as edema and hemorrhage cause increasing pressures in the carpal tunnel, compressing the median nerve. ⁸ Reported lower rates of median nerve symptoms may be due to earlier presentation of patients with these injuries, early reduction of the injury, or differences in the severity of the injuries (eg, Mayfield IV vs II). Although most studies report that carpal tunnel release was performed in patients who presented with nerve symptoms prior to surgery, it has also been demonstrated that the presence of such neuropathy is associated with worse functional outcomes.^9,10 Dorsal fixation can be performed at the time of carpal tunnel release or on an elective basis per physician and patient preference. Carroll et al demonstrated that more than 70% of their patients required carpal tunnel release, whether at index operation or subsequent surgery. They did not comment on the success rate of their closed reductions. ¹¹ However, no factors have yet to be identified that predict which patients are more likely to develop median nerve symptoms, which require operative management, and which self-resolve. ¹⁰ This emphasizes the need to consider early closed reduction attempts for these injuries as opposed to purely awaiting open reduction with definitive surgical treatment.

Weaknesses of this study include the retrospective nature limiting data acquisition such as poor follow-up and recorded patient outcomes. It is unclear whether persistent nerve symptoms were due to failed reduction or the complex injuries making the reduction difficult. This is particularly difficult to analyze for long-term outcomes. Prospective studies with advanced imaging prereduction and postreduction would also offer additional insight into the benefits and risks of attempting reductions with these complex injuries. Perhaps, Herzberg’s suggestion that reduction of high-grade perilunate injuries can lead to damage of otherwise intact ligaments would be supported.

While literature regularly dismisses the use of closed reduction for Herzberg stage IIB perilunate injuries, this study demonstrates high success rate. ¹² This study demonstrates that there remains a place and need for closed reductions in perilunate injuries, including the more severe types, to ensure best possible patient outcomes by protecting neurovascular structures and to assist in triaging access to operating rooms.