The Requirement for Closed Reduction of Dorsally Displaced Unstable Distal Radius Fractures Before Operative Treatment
A randomized controlled study
; ; ; ;
Background: Dorsally displaced distal radius fractures are generally treated with closed reduction followed by casting. Current evidence suggests that fracture reduction is of no benefit before either conservative or surgical treatment. It has not been studied to date whether the degree of pain suffered by the patient during preoperative casting is any different if the fracture is reduced beforehand.
Methods: In a prospective, randomized trial, dorsally displaced unstable distal radius fractures were treated surgically, either with or without prior closed reduction (22 and 25 patients, respectively). The primary endpoint was the difference between the pain score (on the Visual Analog Scale) on day 1 after treatment and the initial pain score on presentation. The secondary endpoints included the clinical and radiological outcome and any damage to the median nerve. Moreover, the Krimmer score (strength, mobility, pain, and function of the wrist joint) an the DASH score (Disability of the Arm, Shoulder and Hand) were determined 3 and 12 months after treatment. This trial has been registered with the number DRKS00010570.
Results: With regard to the primary endpoint on day 1 after treatment, there was a statistically significant non-inferiority of the group without reduction, compared to the group with reduction. Sensory disturbances appeared at similar frequencies in the two groups four to six weeks after treatment (9.5% with reduction, 9.1% without). At 12 months, the Krimmer and DASH scores of patients whose fractures had not been reduced were no worse than those of patients whose fractures had been reduced (96 and 7 versus 96.5 and 4.5, respectively; p-values for non-inferiority, 0.004 and 0.008).
Conclusion: This trial shows that dispensing with closed reduction before casting as a preliminary to planned surgery yields no disadvantage. Thus, in the authors’ view, routine reduction is not warranted.
Distal radius fractures are one of the most common fractures in humans. Due to osteoporosis, the lifetime risk is higher for women (15%) than men (2%) (1). In recent years, distal radius fractures have increasingly been treated operatively (Box 1) (2), even if studies have shown no advantage over nonoperative treatment (3, 4). Unstable displaced fractures are usually treated using palmar fixed-angle plate fixation (5), as the anatomically correct position obtained by reduction is often redisplaced with these bone fractures. Although surgery is planned in the event of instability, it is common to treat fractures with reduction for pain relief prior to cast immobilization (6). Reduction should reduce both bone impaction and the increased tension on soft tissues, especially the median nerve. Even if the fracture morphology indicates operative treatment, reduction is believed to create a better starting condition for nonoperative treatment in the case that surgery is not possible for any reason (7).
However, prospective studies have shown significantly poorer clinical and radiological results when the fractures were reduced prior to cast application (8, 9). In most cases, it can be assumed that the improved, reduced position is completely lost before the bone has healed (10).
For fractures treated operatively, a retrospective study on 1,511 patients showed no negative effects on the clinical or radiological result if closed reduction was not performed in the initial treatment (6).
Reduction initially lessens impaction of the fracture. However, a stable correction of the displacement is unlikely; rather, repositioning of a relatively stable impaction is likely to lead to a painful, unstable situation. In a previous study, a mean pain level of at least 4 on the visual analog scale (VAS) was found after repositioning despite cast immobilization (11).
To date, the degree to which pain during cast application depends on a previously performed reduction has not yet been investigated. The aim of this single-center randomized study was to examine whether not performing closed reduction prior to cast application would negatively affect pain perception in the preoperative period. A further aim of the study was to clarify whether a prolonged time of displacement leads to more frequent occurrence of median nerve complications or to poorer clinical–radiological results.
Primary objective of the study
This non-inferiority study was designed to analyze whether unstable distal radius fractures that were unreduced prior to cast application leads to greater pain perception in the preoperative period. Pain severity was self-assessed by patients using a VAS from 0 (no pain) to 10 (worst possible pain) and documented on a pain assessment forms at the following times:
- prior to starting medical treatment;
- during reduction;
- during cast application;
- the evening following the initial treatment;
- every day following the initial treatment, up to and including the day of operation.
The primary endpoint was the difference in pain perception on the first day after reduction as compared to the initial pain perception. A difference between the two groups on a scale of at least 1 was considered relevant.
Secondary objective of the study
As a secondary criterion, negative outcomes for patients due to not performing reduction were investigated. On the one hand, this concerns any traction damage to the median nerve due to the displacement of the radius, with expected hypoesthesia in the nerve supply area. In order to record this, the patients were asked at four weeks post-operation about tingling sensations and decreased sensitivity. A further criterion for this was the patient‘s subjective information on tingling sensation on the “Disability of the Arm, Shoulder and Hand” (DASH) questionnaire after 3 and 12 months (question 26 of the questionnaire: “Tingling [pins and needles] in your shoulder, arm, or hand”; 1 = none, 5 = extreme).
Further secondary criteria were the functional results of treatment at the follow-up examinations at 4 weeks, 3 months, and 12 months postoperatively. The DASH questionnaire, comprising 30 questions, was used as a purely subjective measuring instrument for the function of the hand and the upper extremity. The scale ranges from 0 to 100, with lower values corresponding to better results (12). The Krimmer score was used as a mixed subjective–objective measuring instrument. It includes questions about pain, ability to function at work and in everyday life, strength (in a side-by-side comparison), and mobility (in extension / flexion, ulnar / radial deviation, and pronation / supination) of the wrist (13). This scale also ranges from 0 to 100, with lower values indicating a worse result. A difference of 10 was considered relevant for both scores.
This prospective randomized trial was carried out at a regional trauma center. The trial protocol was approved by the responsible ethics committee (document number 300/15) and registered in the German Clinical Trials Register (DRKS, Deutsches Register Klinischer Studien) under the number DRKS00010570. Informed consent was obtained from all patients. Patients were enrolled in the period from 16 June 2016 to 31 March 2018.
Inclusion and exclusion criteria
The study included adult men and women with unstable distal radius fractures who presented first to the clinic (Box 2). Suitable patients were informed about the trial background and process; if they agreed, they were included in the study. Patient assignment to one of the two groups was done by randomization using a pre-sorted, sealed envelope method. Information in the envelopes contained not only the group assignment but also the precise course of further treatment. A random allocation sequence was determined at the beginning of the trial using 1:1 block randomization.
Group 1—Plaster cast application after closed reduction of fracture
Patients in this group were kept in axial traction for 15 minutes prior to fluoroscopy-guided closed reduction with intravenous analgesic sedation using midazolam and ketanest. Still under traction, a circular, longitudinally-split, below-elbow plaster cast was applied.
Group 2—Plaster cast application without prior closed reduction of fracture
Patients in this group received a circular, longitudinally split, below-elbow cast without prior reduction. If necessary, intravenous analgesia with novaminsulfon or paracetamol was administered.
Groups 1 and 2
After plaster cast application, the fracture position during cast immoblization was assessed radiologically. Patients completed a pain assessment questionnaire, to assess the level of pain of the previous treatment and on each day following preoperative treatment until the day of surgery. Patients in both groups received ibuprofen and novaminsulfon as prescription medications until surgery. Surgical interventions were planned depending on the availability of operating room capacities. Interventions were carried out using standardized techniques (15). Fractures were treated by open reduction with a palmar approach and stabilized using a fixed-angle plate. The treating surgeon was free to choose between the two available plate systems. Patients received an below-elbow palmar cast for a period of usually two weeks. If there were weak bones or accompanying injuries, cast immobilization could be individually extended to four weeks. Physiotherapy was prescribed after cast treatment.
At four to five weeks postoperatively, patients were clinically and radiologically examined by an independent trauma surgeon who was blinded to the group assignment of patients. At this point, patients were allowed to increase the load on the arm in a manner adapted to the pain. Further physiotherapy was prescribed if necessary.
Patients were clinically examined again at 3 and 12 months after the operation. Mobility was measured with a goniometer, and rough grip strength was measured with a Jamar dynamometer using level 3. The Krimmer wrist score was determined from the measured values and the patient‘s subjective information. In addition, patients were asked to fill out the DASH questionnaire. In the final examination after 12 months, a final radiological examination of the wrist was also carried out in two planes. An assessment of the palmar and ulnar inclinations of the radius, as well as of the ulnar protrusion in the technique according to Gelberman et al. (16), were made. The plate prominence was assessed according to Soong et al. (17).
Sample size calculations were performed using the PASS software version 13 and a t-test for non-inferiority based on data from a pilot study (11). According to this, 19 patients per group should be examined (eBox). Mean differences with a 95% confidence interval (CI) were used to determine the equality of mean values; in accordance with the study protocol, the differences in pain levels during the cast treatment in relation to the level before the start of treatment were initially used. This was to investigate any pain-relieving effects of cast immobilization. A value of 1 on the respective scale was assumed as the relevant non-inferiority limit, corresponding to an otherwise usual equivalence limit of 10%. For the comparison of the absolute pain levels, the limit of 2 recommended in the review by Schomacher (18) was rated as relevant. As seen in the Figure, there was a skewed distribution of the parameters of the clinical–radiological examination, and therefore the Mann-Whitney U-test was used. Median and interquartile ranges (IQR) were given accordingly. For the non-inferiority test of the Krimmer and DASH scores, the non-inferiority limit of 10 was set using one-sided U-test statistics. In accordance with the intention to treat principle, secondary data from patients who left the study were also included in the analysis.
The trial profile is shown in the Figure in accordance with the consolidated standards for reporting trials (CONSORT). Table 1 shows the patient data, which were comparable for both groups.
A loss of reduction was observed for an 85-year-old patient from the reduction group, due to intra-articular movement of screws, and for a 53-year-old patient from the non-reduction group, due to failure of the fixed-angle anchorage with palmar plate prominence; both required early removal of the plates. One patient per group complained of carpal tunnel syndrome symptoms post-trauma; their plates were removed, and a carpal tunnel decompression was carried out at the same time. No infections or other complications, or complex regional pain syndrome, were reported. In ten patients from the reduction group, and twelve patients from the non-reduction group, the plate was removed before or after the final examination.
Pain perception of preoperative treatment
The 22 patients in the reduction group had a lower perception of pain prior to the start of treatment (4.73 ± 2.76 [mean ± standard deviation]) than the 25 patients in the non-reduction group (5.84 ± 2.75) (Table 2). After the analgesic sedation, these patients reported mean pain scores of 1.50 ± 2.54 during the reduction maneuver, and of 1.14 ± 1.94 during the plaster cast application. In the non-reduction group, analgesic sedation was mostly waived, which is why pain scores of 4.24 ± 2.20 persisted during the plaster cast application. As the initial pain level in the non-reduction group was higher, the mean pain in this group was higher than in the reduction group at all times after the plaster cast application and until the operation. On days 1 and 2, the pain level in the non-reduction group showed non-inferiority to the pain level in the reduction group, while on the evening after the plaster cast application and on days 3 to 5, the equality of the groups could not be addressed due to the different initial values. With respect to the pain-relieving effect of the cast treatment (Table 2; lower part; difference between pain at time “X” and pain prior to treatment start), the non-reduction group had a slight advantage on days 1 to 3. On day 4, however, there was a slight advantage for the reduction group. Pain relief from the cast application without reduction showed significant non-inferiority on day 1 (primary endpoint), with −1.5 ± 3.0 as compared to −0.7 ± 2.5 for the reduction group. There was also no inferiority for day 2; equality at the other time points could not be addressed.
Secondary criterion: sensitivity disorders
Four weeks after the procedure, two patients in both groups reported sensitivity disorders in the median-nerve innervated area (Table 3). After three months, 19 patients from the non-reduction group, and 17 patients from the reduction group, reported mild or no tingling sensations. Four patients in each group reported moderate, severe, or extreme tingling sensations. After twelve months, 22 patients from the non-reduction group reported mild or no tingling, and only one patient reported a moderate tingling sensation; in the reduction group, 14 patients reported mild or no tingling sensations, and four patients reported moderate tingling sensations.
Secondary criterion: clinical–radiological results
In the reduction group, the postoperative straightening of the palmar inclination was 3° better (p = 0.029). In contrast, the ulnar protrusion in this group was 0.45 mm longer (p = 0.044), and the ulnar inclination of the radius was almost the same. After twelve months, none of the radiological parameters differed between the groups (Table 3). The 9° better extension (p = 0.008) observed in the reduction group at four weeks postoperatively was almost completely balanced after only three months. Strength was 13 percentage points better in the reduction group at three months (p = 0.048) but no longer differed between the two groups at the final examination time point. After three months, the median Krimmer and DASH scores were 73 (95% CI: [62; 87]) and 18 (7; 35), respectively, for the non-reduction group, and 85 (73; 93.5) and 19 (6.5; 35), respectively, for the reduction group. After twelve months, the Krimmer and DASH scores of 96 (85; 100) and 7 (0; 22.25), respectively, for the non-reduction group were non-inferior to the the Krimmer and DASH scores of 96.5 (84.5; 100) and 4.5 (0.75; 31.25), respectively, for the reduction group (p = 0.004 for Krimmer, and p = 0.008 for DASH) (eFigure).
In the present study, the pain perception during preoperative cast application was independent of having had a prior closed reduction or not. On day 1 (primary endpoint) after the initial treatment, we observed not only a statistically significant non-inferiority but also a somewhat more effective pain relief for the group without reduction prior to cast application. This was also observed on day 2. The non-reduction group still hadslightly better pain relief on day 3. However, as 14 of the 47 patients originally included in the study had already been operated on at this time point, non-inferiority could no longer be statistically addressed. The reduction itself was usually not very painful under analgesic sedation. Individual patients nevertheless found the repositioning maneuver extremely painful (VAS of up to 10), so that in the future even more attention should be given to the consistent use of effective analgesic sedation. The mean pain score of 1.50 during the repositioning maneuver in this study, as compared with the 7.47 in the previous study (11), highlights the importance of regular quality controls, even for standardized procedures in the clinic. Along these lines, it is striking that patients in both groups perceived only minimally less pain on the evening after the initial treatment than prior to presenting to the clinic. As pain relief is one of the primary concerns of a newly injured person, the use of analgesics must be urgently improved when providing first aid to patients.
The present study did not find any adverse effects of not performing reduction prior to cast application on the median nerve in terms of traction damage, provided that the displacement of the fracture had not led directly to median symptoms.
Analogous to the previous extensive retrospective study (6), the prospective randomized trial presented here shows no disadvantage of not performing reduction prior to cast application if the fracture will be treated operatively in any case.
Currently, unstable distal radius fractures are preferably treated operatively, using fixed-angle plate osteosynthesis. Although it is common to perform a closed reduction prior to cast application, not performing it had no disadvantages for either pain relief during the preoperative cast treatment or for the clinical–radiological healing progress after one year. Therefore, in the absence of initial median nerve symptoms, a primary closed reduction should no longer be performed prior to operative therapy.
Data sharing statement
Anonymized data, including metadata, as well as the study protocol can be requested up to ten years after publication from the first author, if these are relevant for further analyses within larger cohorts.
Conflict of interest statement
PD Dr. Spies has received meeting participation fees and reimbursement of travel expenses from Medartis AG, Basel, Switzerland.
Prof. Unglaub has received meeting participation fees, reimbursement of travel expenses, and speaking honoraria from IBRA, Basel, Switzerland.
The remaining authors declare that no conflict of interests exists.
Manuscript received on 24 January 2020, revised version accepted on
9 May 2020
Translated from the original German by Veronica A Raker, PhD
PD Dr. med. Steffen Löw
Clinic for Orthopaedic and Trauma Surgery
97980 Bad Mergentheim, Germany
Cite this as:
Löw S, Papay M, Spies CK, Unglaub F, Eingartner C:
The requirement for closed reduction of dorsally displaced unstable distal radius fractures before operative treatment—a randomized controlled study.
Dtsch Arztebl Int 2020; 117: 783–9. DOI: 10.3238/arztebl.2020.0783
Clinic for Orthopedic and Trauma Surgery, Caritas Hospital, Bad Mergentheim, Germany: Marion Papay, Prof. Dr. med. Christoph Eingartner
Department of Hand Surgery, Vulpius Hospital, Bad Rappenau, Germany: PD Dr. med. Christian Karl Spies, Prof. Dr. med. Frank Unglaub
Medical Faculty Mannheim, Mannheim, Germany: Prof. Dr. med. Frank Unglaub
|1.||Nguyen ND, Ahlborg HG, Center JR, Eisman JA, Nguyen TV: Residual lifetime risk of fractures in women and men. J Bone Miner Res 2007; 22: 781–8 CrossRef MEDLINE|
|2.||Azad A, Kang HP, Alluri RK, Vakhshori V, Kay HF, Ghiassi A: Epidemiological and treatment trends of distal radius fractures across multiple age groups. J Wrist Surg 2019; 8: 305–11 CrossRef MEDLINE PubMed Central|
|3.||Arora R, Lutz M, Deml C, Krappinger D, Haug L, Gabl M: A prospective randomized trial comparing nonoperative treatment with volar locking plate fixation for displaced and unstable distal radial fractures in patients sixty-five years of age and older. J Bone Joint Surg Am 2011; 93: 2146–53 CrossRef MEDLINE|
|4.||Bartl C, Stengel D, Bruckner T, Gebhard F and the ORCHID Study Group: The treatment of displaced intra-articular distal radius fractures in elderly patients—a randomized multi-center study (ORCHID) of open reduction and volar locking plate fixation versus closed reduction and cast immobilization. Dtsch Arztebl Int 2014; 111: 779–87. VOLLTEXT|
|5.||de Putter CE, Selles RW, Polinder S, et al.: Epidemiology and health-care utilisation of wrist fractures in older adults in The Netherlands, 1997–2009. Injury 2013; 44: 421–6 CrossRef MEDLINE|
|6.||Teunis T, Mulder F, Nota SP, Milne LW, Dyer GS, Ring D: No difference in adverse events between surgically treated reduced and unreduced distal radius fractures. J Orthop Trauma 2015; 29: 521–5 CrossRef MEDLINE|
|7.||Wichlas F, Haas NP, Lindner T, Tsitsilonis S: Closed reduction of distal radius fractures: does instability mean irreducibility? Arch Orthop Trauma Surg 2013; 133: 1073–8 CrossRef MEDLINE|
|8.||Neidenbach P, Audigé L, Wilhelmi-Mock M, Hanson B, De Boer P: The efficacy of closed reduction in displaced distal radius fractures. Injury 2010; 41: 592–8 CrossRef MEDLINE|
|9.||Brogren E, Petranek M, Atroshi I: Cast-treated distal radius fractures: a prospective cohort study of radiological outcomes and their association with impaired calcaneal bone mineral density. Arch Orthop Trauma Surg 2015; 135: 927–33 CrossRef MEDLINE|
|10.||Kelly AJ, Warwick D, Crichlow TP, Bannister GC: Is manipulation of moderately displaced Colles’ fracture worthwhile? A prospective randomized trial. Injury 1997; 28: 283–7 CrossRef|
|11.||Löw S, Papay M, Eingartner C: Pain perception following initial closed reduction in the preoperative care of unstable, dorsally displaced distal radius fractures. J Hand Microsurg 2019; 11: 111–6 CrossRef MEDLINE PubMed Central|
|12.||Gummesson C, Atroshi I, Ekdahl C: The disabilities of the arm, shoulder and hand (DASH) outcome questionnaire: longitudinal construct validity and measuring self-rated health change after surgery. BMC Musculoskelet Disord 2003; 4: 1 CrossRef MEDLINE PubMed Central|
|13.||Kalb K, Ludwig A, Tauscher A, Landsleitner B, Wiemer P, Krimmer H: Treatment outcome after surgical arthrodesis. Handchir Mikrochir Plast Chir 1999; 31: 253–9 CrossRef MEDLINE|
|14.||Lafontaine M, Hardy D, Delince P: Stability assessment of distal radius fractures. Injury 1989; 20: 208–10 CrossRef|
|15.||Löw S, Herold D, Eingartner C: Standardized palmar plating of dorsally displaced distal radius fractures. Tech Hand Up Extremity Surg 2013; 17: 106–11 CrossRef MEDLINE|
|16.||Gelberman RH, Salamon PB, Jurist JM, Posch JL: Ulnar variance in Kienböck’s disease. J Bone Joint Surg Am 1975; 57: 674–6 CrossRef MEDLINE|
|17.||Soong M, Earp BE, Bishop G, Leung A, Blazar P: Volar locking plate implant prominence and flexor tendon rupture. J Bone Joint Surg Am 2011; 93: 328–35 CrossRef MEDLINE|
|18.||Schomacher J: Quality criteria of the Visual Analogue Scale for Pain Assessment. Physioscience 2008; 4: 125–33 CrossRef|