J Shoulder Elbow Surg (2014) 23, 173-181
www.elsevier.com/locate/ymse
SHOULDER
Operative versus nonoperative treatment in the management of midshaft clavicular fractures: a meta-analysis of randomized controlled trials Jing Xu, MD, Lei Xu, MD, PhD, Wendong Xu, MD, PhD, Yudong Gu, MD, Jianguang Xu, MD, PhD* Department of Hand Surgery, Huashan Hospital, Fudan University, Shanghai, China Background: There is no consensus on the effects of operative versus nonoperative treatment on the outcomes of midshaft clavicular fractures in adults. We conducted a meta-analysis of randomized clinical studies. Materials and methods: We searched the literature and included studies that investigated the effects of operative versus nonoperative intervention on the outcome of midshaft clavicular fractures. Patient data were pooled by use of standard meta-analytic approaches. For the continuous variables, the weighted mean difference was used. For dichotomous data, the relative risk was calculated. Results: Seven studies reported in 8 publications were eligible for data extraction. The pooled analyses showed that, compared with nonoperative treatment, operative treatment led to significantly lower incidences of nonunion and fewer symptomatic malunions. Subgroup analysis indicated that these advantages could be ascribed to plate fixation. Furthermore, surgery with plates resulted in significantly fewer complications. Patients undergoing surgery had better Disabilities of the Arm, Shoulder and Hand and Constant scores and lower dissatisfaction with their appearance. Conclusion: In the management of midshaft clavicular fractures, surgery is superior to nonoperative treatment. Surgery with plates results in lower incidences of nonunion, fewer total complications, and fewer symptomatic malunions compared with nonoperative treatment. Level of evidence: Level II, Meta-Analysis. Ó 2014 Journal of Shoulder and Elbow Surgery Board of Trustees. Keywords: Midshaft clavicular; plating; intramedullary fixation; nonoperative treatment; meta-analysis
Traditionally, midshaft clavicular fractures have been managed nonoperatively, even when substantially displaced.7 Recent studies have shown a greater prevalence of nonunion, symptomatic malunion, and poor functional outcomes after nonsurgical management of displaced fractures. Although midshaft clavicular fractures have
always been managed conservatively, surgery is becoming increasingly accepted as the optimal treatment method for displaced midshaft clavicular fractures.2,10-12 A few randomized controlled trials (RCTs) concerning surgical versus nonoperative treatment have been published in recent years. However, the relatively small sample size in
Institutional review board or ethical committee approval was not required for this study. This work was supported by funding from the National Natural Science Foundation of China (grant 30901521).
*Reprint requests: Jianguang Xu, MD, PhD, Department of Hand Surgery, Huashan Hospital, Fudan University, No. 12, Wulumuqi Zhong Road, Shanghai, 200040 China. E-mail address: [email protected] (J. Xu).
1058-2746/$ - see front matter Ó 2014 Journal of Shoulder and Elbow Surgery Board of Trustees. http://dx.doi.org/10.1016/j.jse.2013.06.025
174 each published study made the results inconclusive and controversial. Recently, a meta-analysis of RCTs compared operative versus nonoperative approaches for the treatment of midshaft clavicular fractures.14 Regrettably, analysis of publication bias, which is an essential part of a metaanalysis, was not performed in that study, thereby making the conclusions questionable. Another meta-analysis, without assessment of publication bias, only included 4 RCTs.24 Moreover, additional RCTs have been published since these earlier meta-analyses,15,21 which makes the present meta-analysis a more precise estimation.
Materials and methods Search strategy This meta-analysis was carried out following the PRISMA (Preferred Reporting Items for Systematic Reviews and MetaAnalyses) Statement, which was established to help authors report a wide array of systematic reviews to assess the benefits and disadvantages of health care interventions.13 We performed a literature search without language restrictions on July 22, 2012, and an updated literature search was performed on February 12, 2013, using the phrase ‘‘clavicular fractures’’ with the limits ‘‘randomized controlled trial.’’ A second search was performed using the phrase ‘‘clavicle fractures’’ with the limits ‘‘randomized controlled trial’’ using PubMed (1949-2013), Ovid’s Medline (1946-2013), and Medline’s in-process and other non-indexed citations (updated up to February 12, 2013), as well as Web of Knowledge and Embase (1966-2013). Further searches using the same keywords and limitations did not identify any additional references. We also conducted a search of the Cochrane Central Register of Controlled Trials. Reference lists of review articles were scanned to find additional publications. In addition, reference lists of all primary articles and previously published systematic reviews and meta-analyses were manually searched for additional eligible studies. Duplicates were removed. Information was carefully extracted from all eligible publications independently by 2 reviewers (J.X. [first author] and L.X.); disagreements were resolved by discussion between them. If a consensus could not be reached, a third investigator (W.X.) adjudicated the disagreements. The search results were then screened based on the following inclusion criteria: (1) the studies had to be RCTs on patients with midshaft clavicular fractures, (2) the studies had to compare operative with nonoperative treatment, and (3) the patients were aged at least 16 years. Exclusion criteria included (1) nonrandomized trials, (2) studies concerning adolescent fractures, and (3) fracture in the proximal or distal third of the clavicle. The Jadad scale was used to assess the quality of included RCTs, where a score of less than 3 indicates low quality.8
Statistics The primary outcome of our analysis was the incidence of nonunion, which is determined using radiographs or, in some cases, with additional computed tomography scans. The secondary outcome was the functional outcome, measured with the Disabilities of the
J. Xu et al. Arm, Shoulder and Hand (DASH) score and the Constant score. Furthermore, data on complications and patient dissatisfaction were collected. We attempted to contact the authors of the studies included to obtain missing information. For studies that did not present standard deviations, the standard deviations were calculated from the P value or confidence interval (CI) following the guidance of the Cochrane Handbook for Systematic Reviews of Interventions.6 For the meta-analysis of continuous variables, the weighted mean difference with 95% CI was used. For dichotomous variables, the relative treatment effect was expressed as relative risk (RR) with 95% CI.4 Statistical heterogeneity was investigated with the c2 test and quantified with the I2 statistic. We anticipated the presence of clinical heterogeneity based on the findings that the fixation methods and implants used in surgery varied among the RCTs. Because the test for heterogeneity had low statistical power, we assumed the presence of heterogeneity a priori and used the random-effects model in all the analyses. Subgroup analyses of plate or intramedullary fixation were conducted when possible. A sensitivity analysis was performed by detecting the effect of each individual study on the pooled effect size. Funnel plots and Egger tests were used to assess possible publication bias. A funnel plot is a simple scatter plot of the intervention effect estimates from individual studies against some measure of each study’s size or precision. It assumes that the largest studies will be near the mean and smaller studies will be spread on both sides of the mean. Variation from this assumption can indicate publication bias. P < .05 was considered statistically significant. Analyses were performed with the Stata/SE 10.0 program (StataCorp, College Station, TX, USA).
Results Selected studies and characteristics Potentially relevant citations were identified and screened, of which only 6 published RCTs3,9,15,19,21,22 and 1 abstract20 met the inclusion criteria and were selected for this meta-analysis (Fig. 1 and Table I). One study17 provided 2-year follow-up data of the study by the Canadian Orthopaedic Trauma Society.3 In this analysis, the 2 publications were considered as reports of 1 study and were combined. Among the RCTs, 4 compared plate fixation with nonoperative treatment3,15,20,21 and 3 compared intramedullary fixation with nonoperative treatment.9,19,22 The level of evidence for each article was graded with a score from 1 to 3 according to the Jadad score.8 A total of 471 patients were included in the analysis. Sample sizes of the studies ranged from 50 to 111 patients. Of the patients, 231 were randomized to receive surgery and 240 to nonoperative treatment. Among individuals treated surgically, 146 were treated with plate fixation and 94 with intramedullary fixation. One study was a multicenter RCT,3 and the others were single-center RCTs.2,9,19-22 Allocation concealment was reported in 5 trials3,9,15,19,21 and not stated in the other trials.20,22 Because of the obvious nature of the intervention, no trials were double blind.
Operation versus non-operation in clavicular fractures
Figure 1
175
Flowchart of meta-analysis. CENTRAL, Cochrane Central Register of Controlled Trials.
Effects of operative versus nonoperative management on incidence of nonunion All 7 studies reported nonunion incidence. The incidence of nonunion in both the operative and nonoperative groups reported by Witzel22 was zero, so this study was excluded when we performed the analysis. The pooled results of our primary outcome measure, nonunion incidence, presented a statistically significant difference favoring operative over nonoperative treatment (RR, 0.25; 95% CI, 0.100.67; P < .01). Subgroup analysis concerning fixation methods showed that plate fixation (RR, 0.22; 95% CI, 0.07-0.66; P < .01), but not intramedullary fixation (RR, 0.42; 95% CI, 0.06-3.04; P ¼ .39), was associated with a reduced risk in comparison with nonoperative treatment (Fig. 2).
Effects of operative versus nonoperative management on functional scores DASH data at 3, 6, and 12 months postoperatively are available from 2 or 3 studies from the total 7 studies (Table II). Only Schemitsch et al17 reported DASH scores at 24 months, which showed that DASH scores in patients in the operative group were superior to those in the nonoperative group, so we were not able to perform a pooled analysis. For Constant scores, the pooled analysis based on 2 or 3 studies indicated
that operative treatment was superior to nonoperative treatment at 6, 12, and 24 months postoperatively but not at 3 months postoperatively (Table II). Because of the limited number of studies, we could not perform subgroup analyses comparing surgery using plates or intramedullary implants with nonoperative treatment.
Effects of operative versus nonoperative management on incidence of complications Pooled results from 6 studies showed that the total complication rate after surgery was not significantly different in comparison with nonoperative treatment (RR, 0.78; 95% CI, 0.56-1.07; P ¼ .12). To further explore the effects of different surgical methods in comparison with nonoperative treatment, subgroup analyses indicated that the total number of complications after surgery using plates was less than that in nonoperative cases (RR, 0.70; 95% CI, 0.54-0.91; P < .01), whereas no significant difference in the total complication rate was detected between surgery using intramedullary fixation and nonoperative treatment (RR, 1.78; 95% CI, 0.28-11.46; P ¼ .55) (Fig. 3). A statistically significant difference in symptomatic malunion was found in favor of surgical treatment. Further subgroup analysis showed that surgery with plates but not intramedullary fixation was superior to nonoperative treatment. Regarding the incidence of further surgical
176
Table I
Demographic characteristics of clinical trials included in analysis Operative treatment
Nonoperative treatment
Follow-up time (mo)
Outcomes measured
Jadad score8
4/4
AO/ASIF stainless steel reconstruction plates
Sling
12
3
26/24
20/21
3.5-mm reconstruction plates
Sling and elastic cotton band
12
18-65
30/30
4/4
Elastic stable intramedullary nails
Sling
24
Judd et al,9 2009
17-40
29/28
2/3
Modified Hagie pins
Sling
12
COTS,3,17 2007
16-60
62/49
9/15
Limited-contact dynamic compression plates, reconstruction plates, precontoured plates, other plates
Sling
12
Witzel,22 2007
18-72
35/33
13/10
Intramedullary titanium pins
Backpack association
8-26
Smith et al,20 2001
Adults
30/35
d
Small-fragment plates
Sling
Mean, 18.5
Nonunion, DASH and Constant scores, pain, fracture healing, and complications Nonunion, DASH and Constant scores, patient satisfaction, range of motion of shoulder, and complications Nonunion, DASH and Constant scores, patient satisfaction, and complications Nonunion, SANE and L’Insalata functional scores, and complications Nonunion, DASH and Constant scores, patient satisfaction, range of motion, complications, and appearance of shoulder Nonunion, range of motion, pain, radiologic results, force measurement, and resumption of physical activity Nonunion and patient satisfaction
Authors
Age (y)
No. of patients (operative/ nonoperative)
Virtanen et al,21 2012
18-70
28/32
Mirzatolooei,15 2011
18-65
Smekal et al,19 2009
No. of female patients (operative/ nonoperative)
3
2
3
1
1
J. Xu et al.
COTS, Canadian Orthopaedic Trauma Society; SANE, Single Assessment Numeric Evaluation.
2
Operation versus non-operation in clavicular fractures
177
Figure 2 Assessment of effects of operative versus nonoperative treatment on incidence of nonunion. Squares with diamonds indicate weighting given to the trial in the overall pooled estimate, taking into account the number of participants and the amount of interstudy variation (heterogeneity); rhombus, combined effect size. COTS, Canadian Orthopaedic Trauma Society.
Table II
Comparison of operative and nonoperative treatments with respect to outcomes of DASH and Constant scores
Parameter
DASH score
Constant score
Time
3 mo 6 mo 12 mo 24 mo 3 mo 6 mo 12 mo 24 mo
No. of studies
3 2 2 1 2 2 3 2
No. of participants O
N
118 9 114 52 88 92 114 82
104 79 98 43 74 79 98 73
WMD
95% CI
P value
Treatment favored
4.81 3.43 7.90 7.3 3.19 7.75 5.89 4.47
8.90 to 0.63 6.05 to 0.80 13.62 to 2.18 13.49 to 1.11 5.39 to 11.77 4.59 to 10.92 1.14 to 10.64 2.48 to 6.46
.02 .01
www.elsevier.com/locate/ymse
SHOULDER
Operative versus nonoperative treatment in the management of midshaft clavicular fractures: a meta-analysis of randomized controlled trials Jing Xu, MD, Lei Xu, MD, PhD, Wendong Xu, MD, PhD, Yudong Gu, MD, Jianguang Xu, MD, PhD* Department of Hand Surgery, Huashan Hospital, Fudan University, Shanghai, China Background: There is no consensus on the effects of operative versus nonoperative treatment on the outcomes of midshaft clavicular fractures in adults. We conducted a meta-analysis of randomized clinical studies. Materials and methods: We searched the literature and included studies that investigated the effects of operative versus nonoperative intervention on the outcome of midshaft clavicular fractures. Patient data were pooled by use of standard meta-analytic approaches. For the continuous variables, the weighted mean difference was used. For dichotomous data, the relative risk was calculated. Results: Seven studies reported in 8 publications were eligible for data extraction. The pooled analyses showed that, compared with nonoperative treatment, operative treatment led to significantly lower incidences of nonunion and fewer symptomatic malunions. Subgroup analysis indicated that these advantages could be ascribed to plate fixation. Furthermore, surgery with plates resulted in significantly fewer complications. Patients undergoing surgery had better Disabilities of the Arm, Shoulder and Hand and Constant scores and lower dissatisfaction with their appearance. Conclusion: In the management of midshaft clavicular fractures, surgery is superior to nonoperative treatment. Surgery with plates results in lower incidences of nonunion, fewer total complications, and fewer symptomatic malunions compared with nonoperative treatment. Level of evidence: Level II, Meta-Analysis. Ó 2014 Journal of Shoulder and Elbow Surgery Board of Trustees. Keywords: Midshaft clavicular; plating; intramedullary fixation; nonoperative treatment; meta-analysis
Traditionally, midshaft clavicular fractures have been managed nonoperatively, even when substantially displaced.7 Recent studies have shown a greater prevalence of nonunion, symptomatic malunion, and poor functional outcomes after nonsurgical management of displaced fractures. Although midshaft clavicular fractures have
always been managed conservatively, surgery is becoming increasingly accepted as the optimal treatment method for displaced midshaft clavicular fractures.2,10-12 A few randomized controlled trials (RCTs) concerning surgical versus nonoperative treatment have been published in recent years. However, the relatively small sample size in
Institutional review board or ethical committee approval was not required for this study. This work was supported by funding from the National Natural Science Foundation of China (grant 30901521).
*Reprint requests: Jianguang Xu, MD, PhD, Department of Hand Surgery, Huashan Hospital, Fudan University, No. 12, Wulumuqi Zhong Road, Shanghai, 200040 China. E-mail address: [email protected] (J. Xu).
1058-2746/$ - see front matter Ó 2014 Journal of Shoulder and Elbow Surgery Board of Trustees. http://dx.doi.org/10.1016/j.jse.2013.06.025
174 each published study made the results inconclusive and controversial. Recently, a meta-analysis of RCTs compared operative versus nonoperative approaches for the treatment of midshaft clavicular fractures.14 Regrettably, analysis of publication bias, which is an essential part of a metaanalysis, was not performed in that study, thereby making the conclusions questionable. Another meta-analysis, without assessment of publication bias, only included 4 RCTs.24 Moreover, additional RCTs have been published since these earlier meta-analyses,15,21 which makes the present meta-analysis a more precise estimation.
Materials and methods Search strategy This meta-analysis was carried out following the PRISMA (Preferred Reporting Items for Systematic Reviews and MetaAnalyses) Statement, which was established to help authors report a wide array of systematic reviews to assess the benefits and disadvantages of health care interventions.13 We performed a literature search without language restrictions on July 22, 2012, and an updated literature search was performed on February 12, 2013, using the phrase ‘‘clavicular fractures’’ with the limits ‘‘randomized controlled trial.’’ A second search was performed using the phrase ‘‘clavicle fractures’’ with the limits ‘‘randomized controlled trial’’ using PubMed (1949-2013), Ovid’s Medline (1946-2013), and Medline’s in-process and other non-indexed citations (updated up to February 12, 2013), as well as Web of Knowledge and Embase (1966-2013). Further searches using the same keywords and limitations did not identify any additional references. We also conducted a search of the Cochrane Central Register of Controlled Trials. Reference lists of review articles were scanned to find additional publications. In addition, reference lists of all primary articles and previously published systematic reviews and meta-analyses were manually searched for additional eligible studies. Duplicates were removed. Information was carefully extracted from all eligible publications independently by 2 reviewers (J.X. [first author] and L.X.); disagreements were resolved by discussion between them. If a consensus could not be reached, a third investigator (W.X.) adjudicated the disagreements. The search results were then screened based on the following inclusion criteria: (1) the studies had to be RCTs on patients with midshaft clavicular fractures, (2) the studies had to compare operative with nonoperative treatment, and (3) the patients were aged at least 16 years. Exclusion criteria included (1) nonrandomized trials, (2) studies concerning adolescent fractures, and (3) fracture in the proximal or distal third of the clavicle. The Jadad scale was used to assess the quality of included RCTs, where a score of less than 3 indicates low quality.8
Statistics The primary outcome of our analysis was the incidence of nonunion, which is determined using radiographs or, in some cases, with additional computed tomography scans. The secondary outcome was the functional outcome, measured with the Disabilities of the
J. Xu et al. Arm, Shoulder and Hand (DASH) score and the Constant score. Furthermore, data on complications and patient dissatisfaction were collected. We attempted to contact the authors of the studies included to obtain missing information. For studies that did not present standard deviations, the standard deviations were calculated from the P value or confidence interval (CI) following the guidance of the Cochrane Handbook for Systematic Reviews of Interventions.6 For the meta-analysis of continuous variables, the weighted mean difference with 95% CI was used. For dichotomous variables, the relative treatment effect was expressed as relative risk (RR) with 95% CI.4 Statistical heterogeneity was investigated with the c2 test and quantified with the I2 statistic. We anticipated the presence of clinical heterogeneity based on the findings that the fixation methods and implants used in surgery varied among the RCTs. Because the test for heterogeneity had low statistical power, we assumed the presence of heterogeneity a priori and used the random-effects model in all the analyses. Subgroup analyses of plate or intramedullary fixation were conducted when possible. A sensitivity analysis was performed by detecting the effect of each individual study on the pooled effect size. Funnel plots and Egger tests were used to assess possible publication bias. A funnel plot is a simple scatter plot of the intervention effect estimates from individual studies against some measure of each study’s size or precision. It assumes that the largest studies will be near the mean and smaller studies will be spread on both sides of the mean. Variation from this assumption can indicate publication bias. P < .05 was considered statistically significant. Analyses were performed with the Stata/SE 10.0 program (StataCorp, College Station, TX, USA).
Results Selected studies and characteristics Potentially relevant citations were identified and screened, of which only 6 published RCTs3,9,15,19,21,22 and 1 abstract20 met the inclusion criteria and were selected for this meta-analysis (Fig. 1 and Table I). One study17 provided 2-year follow-up data of the study by the Canadian Orthopaedic Trauma Society.3 In this analysis, the 2 publications were considered as reports of 1 study and were combined. Among the RCTs, 4 compared plate fixation with nonoperative treatment3,15,20,21 and 3 compared intramedullary fixation with nonoperative treatment.9,19,22 The level of evidence for each article was graded with a score from 1 to 3 according to the Jadad score.8 A total of 471 patients were included in the analysis. Sample sizes of the studies ranged from 50 to 111 patients. Of the patients, 231 were randomized to receive surgery and 240 to nonoperative treatment. Among individuals treated surgically, 146 were treated with plate fixation and 94 with intramedullary fixation. One study was a multicenter RCT,3 and the others were single-center RCTs.2,9,19-22 Allocation concealment was reported in 5 trials3,9,15,19,21 and not stated in the other trials.20,22 Because of the obvious nature of the intervention, no trials were double blind.
Operation versus non-operation in clavicular fractures
Figure 1
175
Flowchart of meta-analysis. CENTRAL, Cochrane Central Register of Controlled Trials.
Effects of operative versus nonoperative management on incidence of nonunion All 7 studies reported nonunion incidence. The incidence of nonunion in both the operative and nonoperative groups reported by Witzel22 was zero, so this study was excluded when we performed the analysis. The pooled results of our primary outcome measure, nonunion incidence, presented a statistically significant difference favoring operative over nonoperative treatment (RR, 0.25; 95% CI, 0.100.67; P < .01). Subgroup analysis concerning fixation methods showed that plate fixation (RR, 0.22; 95% CI, 0.07-0.66; P < .01), but not intramedullary fixation (RR, 0.42; 95% CI, 0.06-3.04; P ¼ .39), was associated with a reduced risk in comparison with nonoperative treatment (Fig. 2).
Effects of operative versus nonoperative management on functional scores DASH data at 3, 6, and 12 months postoperatively are available from 2 or 3 studies from the total 7 studies (Table II). Only Schemitsch et al17 reported DASH scores at 24 months, which showed that DASH scores in patients in the operative group were superior to those in the nonoperative group, so we were not able to perform a pooled analysis. For Constant scores, the pooled analysis based on 2 or 3 studies indicated
that operative treatment was superior to nonoperative treatment at 6, 12, and 24 months postoperatively but not at 3 months postoperatively (Table II). Because of the limited number of studies, we could not perform subgroup analyses comparing surgery using plates or intramedullary implants with nonoperative treatment.
Effects of operative versus nonoperative management on incidence of complications Pooled results from 6 studies showed that the total complication rate after surgery was not significantly different in comparison with nonoperative treatment (RR, 0.78; 95% CI, 0.56-1.07; P ¼ .12). To further explore the effects of different surgical methods in comparison with nonoperative treatment, subgroup analyses indicated that the total number of complications after surgery using plates was less than that in nonoperative cases (RR, 0.70; 95% CI, 0.54-0.91; P < .01), whereas no significant difference in the total complication rate was detected between surgery using intramedullary fixation and nonoperative treatment (RR, 1.78; 95% CI, 0.28-11.46; P ¼ .55) (Fig. 3). A statistically significant difference in symptomatic malunion was found in favor of surgical treatment. Further subgroup analysis showed that surgery with plates but not intramedullary fixation was superior to nonoperative treatment. Regarding the incidence of further surgical
176
Table I
Demographic characteristics of clinical trials included in analysis Operative treatment
Nonoperative treatment
Follow-up time (mo)
Outcomes measured
Jadad score8
4/4
AO/ASIF stainless steel reconstruction plates
Sling
12
3
26/24
20/21
3.5-mm reconstruction plates
Sling and elastic cotton band
12
18-65
30/30
4/4
Elastic stable intramedullary nails
Sling
24
Judd et al,9 2009
17-40
29/28
2/3
Modified Hagie pins
Sling
12
COTS,3,17 2007
16-60
62/49
9/15
Limited-contact dynamic compression plates, reconstruction plates, precontoured plates, other plates
Sling
12
Witzel,22 2007
18-72
35/33
13/10
Intramedullary titanium pins
Backpack association
8-26
Smith et al,20 2001
Adults
30/35
d
Small-fragment plates
Sling
Mean, 18.5
Nonunion, DASH and Constant scores, pain, fracture healing, and complications Nonunion, DASH and Constant scores, patient satisfaction, range of motion of shoulder, and complications Nonunion, DASH and Constant scores, patient satisfaction, and complications Nonunion, SANE and L’Insalata functional scores, and complications Nonunion, DASH and Constant scores, patient satisfaction, range of motion, complications, and appearance of shoulder Nonunion, range of motion, pain, radiologic results, force measurement, and resumption of physical activity Nonunion and patient satisfaction
Authors
Age (y)
No. of patients (operative/ nonoperative)
Virtanen et al,21 2012
18-70
28/32
Mirzatolooei,15 2011
18-65
Smekal et al,19 2009
No. of female patients (operative/ nonoperative)
3
2
3
1
1
J. Xu et al.
COTS, Canadian Orthopaedic Trauma Society; SANE, Single Assessment Numeric Evaluation.
2
Operation versus non-operation in clavicular fractures
177
Figure 2 Assessment of effects of operative versus nonoperative treatment on incidence of nonunion. Squares with diamonds indicate weighting given to the trial in the overall pooled estimate, taking into account the number of participants and the amount of interstudy variation (heterogeneity); rhombus, combined effect size. COTS, Canadian Orthopaedic Trauma Society.
Table II
Comparison of operative and nonoperative treatments with respect to outcomes of DASH and Constant scores
Parameter
DASH score
Constant score
Time
3 mo 6 mo 12 mo 24 mo 3 mo 6 mo 12 mo 24 mo
No. of studies
3 2 2 1 2 2 3 2
No. of participants O
N
118 9 114 52 88 92 114 82
104 79 98 43 74 79 98 73
WMD
95% CI
P value
Treatment favored
4.81 3.43 7.90 7.3 3.19 7.75 5.89 4.47
8.90 to 0.63 6.05 to 0.80 13.62 to 2.18 13.49 to 1.11 5.39 to 11.77 4.59 to 10.92 1.14 to 10.64 2.48 to 6.46
.02 .01
