论文部分内容阅读
Abstract：Objective To compare the outcome and safety of the minimally invasive dynamic hip screw（MIDHS）technique and the conventional dynamic hip screw（CDHS）technique for intertrochanteric fractures in the elderly.
Method We searched the database of PubMed，EMBASE，the Cochrane Library，CNKI and VIP from inception to August 2015 for relevant randomized controlled trials（RCTs）without language restriction. References of all the selected articles were hand-searched for any additional trials. Two authors independently extracted data from all eligible studies，including study design，participants，interventions，and outcomes（wound size，surgery time，intraoperative blood loss，length of hospital stay，Harris hip score and severe complications）. Data were analyzed using fixed-effects and random-effects models with mean differences and risk ratios for continuous and dichotomous variables，respectively.
Results Eight studies involving 612 patients were identified in this analysis. The meta-analysis showed smaller wound size（WMD，-7.12；95% CI，-7.68，-6.56，respectively），shorter surgery time（WMD，-21.86；95% CI，-36.09，-7.62，respectively），less intraoperative blood loss（WMD = 4.60 mL，95%CI，227.06–36.27 ml），shorter length of hospital stay（WMD，-3.30；95% CI，-3.72，-2.88，respectively），higher Harris hip score（WMD，1.54；95% CI，0.40，2.67，respectively），and fewer severe complications（RR，0.47，95 % CI 0.22，1.00）in patients receiving MIDHS versus those receiving CDHS.
Conclusions The available evidence suggests that MIDHS is superior to CDHS for intertrochanteric fractures in elderly patients.
Introduction
Hip fractures are becoming increasingly common with the aging of the population and approximately half of these fractures occur in the intertrochanteric region. These fractures typically occur in elderly patients and often result in eventual loss of their functional independence[1，2]. Intertrochanteric fractures in the elderly are associated with higher rates of mortality and complications if they fail to mobilize or ambulate early. For a return to pre-injury functioning and activity levels，early operative interventions are recommended for senile femoral intertrochanteric fracture[3]. The dynamic hip screw has become the standard and is currently the most common implant for intertrochanteric fractures[4]. Several prospective randomized controlled trials（RCT）over the previous decade have shown that minimally invasive dynamic hip screw（MIDHS）can reduce intraoperative blood loss，shorten operative time，and relieve postoperative pain[4-6]. No well-designed meta-analysis and systematic review capable of providing high levels of evidence has been conducted. Recently，a systematic review and meta-analysis were conducted to test the theoretical advantages of MIDHS fixation over CDHS fixation for intertrochanteric hip fractures[7]. However the paper also included quasi- and non-randomized controlled trials（RCT），which reduces the confidence level of the meta-analysis. Therefore，we carried out this meta-analysis to evaluate and compare the clinical outcomes and safety of MIDHS vs. CDHS from published RCTs.
Materials and methods
Search Strategies
We searched the database of PubMed，EMBASE，Cochrane Library，CNKI and VIP for all articles on MIDHS and CDHS for intertrochanteric hip fracture without language restriction. The following search terms were used：（‘intertrochanteric hip fracture’ OR ‘fractures’）AND（randomized controlled trial）. The search was performed in August，2015 and covered literatures published between inception and August，2015. We also retrieved relevant articles in Google Scholar. The bibliographies of the obtained literature and the references of relevant meta-analyses were checked to ensure that no related studies were inadvertently neglected. This meta-analysis was performed in accordance with the PRISMA guidelines[8].
Selection criteria
The inclusion criteria were as follows：1）patients had intertrochanteric fractures，and 2）studies compared MIDHS with CDHS. When two studies were reported by the same institution and authors，the study with the higher quality or the most recent publication was included in the analysis，unless the study outcomes were mutually exclusive or measured at different time points. Exclusion criteria were as follows：1）data collection was not scientific and data analysis was incorrect or not provided，and 2）review literature，repeated reports and retrospective studies.
Data extraction
All data were extracted independently by two authors according to the inclusion criteria. Disagreements were resolved by discussion between the two reviewers. The following characteristics were collected from each study：the first author，year of publication，source，experimental design，sample size，sample characteristics，and primary and secondary outcomes. Un-weighted κ statistics was applied to assess agreement between the two reviewers. The quality of the included studies was assessed using the Newcastle-Ottawa Scale.
Statistical analysis Statistical analysis was conducted using Review Manager 5.2 software. Continuous data were expressed as the weighted mean difference（WMD）and 95 % confidence intervals（CI）. Dichotomous data were presented as risk ratios（RR）with 95 % CI. P ≤ 0.05 was considered statistically significant. Heterogeneity was tested using the chi-square test and the I2 test. A significance level of 0.1 for the chi-square test was interpreted as evidence of heterogeneity. The I2 test was used to estimate the total variation among the studies. When no statistical evidence of heterogeneity was present，a fixed effects model was adopted. If not，a random effects model was chosen.
Results
Characteristics of the included studies
The study flowchart is shown in Fig. 1. Three hundred and twelve publications were identified for full text assessment with the search strategy（91 from PubMed，33 from the Cochrane Library，79 from Embase，42 from CNKI，62 from WanFang and 5 from the references of previous meta analyses）. After exclusion of duplications using Notexpress software and manual confirmation，214 publications without duplications remained. The two reviewers further read the articles and finally 8 papers were included[4-6，9-13].The quality of the included studies was using the Newcastle-Ottawa Scale for all 8 studies. There was excellent agreement between investigators for full text screening（κ= 0.92）. The basic characteristics of these studies such as the authors，publication year，journal，study type，the number of patients，treatment method and follow-up duration are summarized in Table 1. The number of patients in each study ranged from 53 to132 and the total number of patients was 612，including 302 patients receiving MIDHS and 310 patients receiving CDHS. The methodological quality of included studies is presented in Fig. 2.
Wound size
Five RCTs[6，9，10，12，13]studied wound size after MIDHS and CDHS for intertrochanteric fractures. A total of 430 patients were studied，including 204 patients receiving MIDHS and 226 patients receiving CDHS. No heterogeneity was present among the studies（P=0.07，I2=53 %）. There was a statistically significant difference in wound size between MIDHS and CDHS for intertrochanteric fractures（WMD，-7.12；95% CI，-7.68，-6.56，respectively），indicating that MIDHS can significantly reduce wound size（Fig. 3）.
Operative time
Five RCTs [6，9-12]reported the duration of surgery using MIDHS or CDHS for intertrochanteric fractures. MIDHS was used for 209 patients and 227 patients for CDHS. Heterogeneity tests indicated statistical evidence of heterogeneity（P<0.00001，I2=96 %）. Therefore，a random effects model was used. The meta-analysis showed a significant difference in operative time between MIDHS and CDHS for intertrochanteric fractures（WMD，-21.86；95% CI，-36.09，-7.62），indicating MIDHS treatment requires less surgery time（Fig. 4）. Intraoperative blood loss
Four studies[6，10，12，13]reported data on intraoperative blood loss. Due to high heterogeneity（P < 0.00001，I2 = 92%），a random-effects model was adopted to pool the data. There was a significant difference between the two groups regarding intraoperative blood loss（WMD = -202.99 ml，95% CI，-221.57，–81.80 mL；P = 0.25）（Fig. 5）. A further sensitivity analysis，which was performed after one RCT15was excluded，indicated that no significant difference（WMD = 4.60 ml，95%CI，227.06–36.27 mL；P = 0.78；fixed-effects model）with low heterogeneity（P = 0.31，I2 = 14%）between the two groups（Fig. 5）.
Length of hospital stay
Three RCTs[9，11，12] reported the length of hospital stay after MIDHS and CDHS for intertrochanteric fractures. One hundred forty-nine patients were included in the MIDHS group and 161 patients in the CDHS group. No heterogeneity was observed between the studies（P= 0.64，I2 = 0 %）. Therefore，a fixed effects model was used. Our meta-analysis showed a statistical difference in the length of hospital stay between MIDHS and CDHS for intertrochanteric fractures（WMD，-3.30；95% CI，-3.72，-2.88）. MIDHS significantly reduced the length of hospital stay（Fig. 6）.
Harris hip score
Five RCTs [5，6，9，11，13] reported Harris hip scores after MIDHS and CDHS treatment of intertrochanteric fractures. One hundred forty-nine patients were included in the MIDHS group and 16 patients in the CDHS group. No heterogeneity was present between the studies（P = 0.64，I2 = 0 %）.Therefore，a fixed effects model was used. The meta-analysis demonstrated a statistically significant difference in Harris hip scores between MIDHS and CDHS for intertrochanteric fractures（WMD，1.54；95% CI，0.40，2.67）. The postoperative Harris hip score was significantly higher in the MDHS group（Fig. 7）.
Severe complications
Six RCTs[4，6，9-11，13]provided data on severe complication rate after MIDHS and CDHS treatment for intertrochanteric fractures. A total of 447 patients were studied，including 213 patients receiving MIDHS treatment and 234 patients receiving CDHS treatment. No heterogeneity was present between the studies（P=0.05，I2 =3 %）. The result of meta-analysis showed a statistical difference in the neurological complication rate between the MIDHS group and the CDHS group（RR 0.47，95 %）（Fig. 8）. In the CDHS group，6 patients developed deep vein thrombosis（DVT）and 2 patients had pulmonary embolism，6 patients had implant failure，1 patient developed atrial fibrillation，and 1 patient had urinary tract infection. In addition，5 patients died. In the MIDHS group，4 patients developed DVT，and 2 paitents had implant failure. Two patients died in the group. Discussion
The treatment of intertrochanteric femoral fractures remains a challenge for orthopedic surgeons due to lack of unified standard in fracture classification，treatment，and curative effect evaluation. Minimally invasive surgery has gained in popularity in orthopedic surgery as it is associated with reduced operation time，smaller blood loss and decreased postoperative pain，reduced postoperative morbidity，and faster recovery of function. The purpose of this meta-analysis and systematic review was to summarize the currently available evidence to determine the safety and efficacy of the minimally invasive technique，MIDHS，vs. CDHS for intertrochanteric femoral fractures in the elderly. This meta-analysis indicates that MIDHS and CDHS differ significantly in wound size，operative time，intraoperative blood loss，length of hospital stay，Harris hip score，and severe complications.
Intertrochanteric femoral fractures in the elderly are associated with higher rates of co-morbidities such as DVT，urinary tract infections，and pulmonary embolism[14]. A reduced operative time，especially in elderly patients with comorbid conditions or poor cardiopulmonary reserve，is desirable because it reduces temporal exposure to the risks of general anesthesia. This，combined with reduced surgical trauma，may be significant in reducing postoperative morbidity and mortality in such patients. Intraoperative blood loss in MIDHS was reduced significantly compared with CDHS. This can be attributed to smaller incisions，less soft-tissue damage，and minimally operative trauma without exposing the fracture in MIDHS. Smaller intraoperative blood loss reduces the need for and avoids the hazards of blood transfusion for elderly patients[15-17]，thus helping minimize the healthcare cost for intertrochanteric hip fracture，which imposes a mounting economic burden on the patients，the family and the society.
Early mobilization largely depends on the reduction of postoperative pain，and some studies show a trend toward lower pain scores in patients receiving MIDHS [4，11].which may encourage earlier ambulation and，theoretically，facilitate earlier discharge. MIDHS requires less operative time，reduces the amount of intraoperative blood loss，and allows patients to be discharged sooner，which may benefit patients’ recoveries and has significant financial implications for hospitals. The rate of severe postoperative complications in the MIDHS group（3.8%）was markedly lower than that of the CDHS group（8.5%）. Elderly patients are at higher risks for DVT，urinary tract infections，and pulmonary embolism if they fail to mobilize or ambulate early. Reduced operative time，less bleeding，and less postoperative pain may promote earlier ambulation in patients receiving MIDHS，which may contribute to a more favorable outcome in terms of severe postoperative complications. The implant failure rates are similar between patients receiving MIDHS and those receiving CDHS，possibly because they are unstable fractures with loss of medial calcar continuity and with osteoporosis，and bone grafting is not done intraoperatively[6，10]. The MIDHS is safer and more dependable than the CDHS because of severe postoperative complications and should be the first option for treatment of intertrochanteric fractures. Harris hip score is a multidimensional observational assessment of pain，walking function，activities of daily living，and hip joint range of motion[18]. Our meta-analysis showed significantly higher scores of patients receiving MIDHS than those receiving CHDS，implying a faster functional recovery of patients receiving MIDHS. Because intertrochanteric fractures usually occur in elderly patients who may have age-associated cognitive deficits or other diseases，it is important for elderly patients to return to pre-injury activity levels as soon as possible to avoid complications. Higher Harris hip scores of the MIDHS group suggest MIDHS may be more beneficial for elderly patients.
Our study focus on systematic reviews of RCTs，which are likely to provide more reliable information than other sources of evidence on the differential effects of alternative forms of healthcare[19]. Undoubtedly，this analysis also has some limitations. Significant heterogeneity was observed between the included trials for intraoperative blood loss，and operative time. This heterogeneity may be attributable to variation in the skills of the surgeons and the different types of peritrochanteric fractures. The eligibility criteria for inclusion of patients with peritrochanteric fractures differ from one another，which might influence consistency of effects across the included studies and cause between-study heterogeneity. To ensure uniformity in both defining patients’ characteristics for peritrochanteric fractures and defining outcome measures，an individual patient data meta-analysis is needed[20]. Moreover，we did not undertake a subgroup analysis of the different fracture types to identify the possible source of heterogeneity because not all of the included studies described data according to different fracture types. Furthermore，the effects might differ by different fracture types，which require further study.
Conclusions
Based on this systematic review and meta-analysis，MIDHS is superior to CHDS in terms of wound size，surgery time，intraoperative blood loss，length of hospital stay，Harris hip score and severe complications. Thus，the MIDHS technique is recommended as a minimally invasive technique that can be considered as an additional alternative treatment for intertrochanteric fractures，especially in elderly patients with multiple morbidities.
Conflict of interest
The authors declare that they have no conflict of interest.
Figure legends Figure 1 The study flowchart.
Figure 2 Methodological quality of included studies. The risk of bias tool incorporates assessment of randomization（sequence generation and allocation concealment），blinding（participants，personnel，and outcome assessors），completeness of outcome data，selection of outcomes reported，and other sources of bias. The items are scored with “yes，” “no，” and “unclear.”
Figure 3 Comparison of wound size（cm）between the MIDHS and CDHS group. Key relative risk（square）and its 95% confidence interval（CI）（horizontal line）；pooled evidence（large black diamond）；relative risk of 1.0（vertical black line）.
Figure 4 Comparison of operative time（minutes）between the MIDHS and CDHS treatments. Key relative risk（square）and its 95% confidence interval（CI）（horizontal line）；pooled evidence（large black diamond）；relative risk of 1.0（vertical black line）.
Figure 5 Comparison of intraoperative blood loss（mL）between the MIDHS and CDHS group. Key relative risk（square）and its 95% confidence interval（CI）（horizontal line）；pooled evidence（large black diamond）；relative risk of 1.0（vertical black line）.
Figure 6 Comparison of length of hospital stay（days）between the MIDHS and CDHS group. Key relative risk（square）and its 95% confidence interval（CI）（horizontal line）；pooled evidence（large black diamond）；relative risk of 1.0（vertical black line）.
Figure 7 Comparison of postoperative Harris hip score between the MIDHS and CDHS group. Key relative risk（square）and its 95% confidence interval（CI）（horizontal line）；pooled evidence（large black diamond）；relative risk of 1.0（vertical black line）.
Figure 8 Comparison of serious postoperative complication rates between the MIDHS and CDHS treatments. Key relative risk（square）and its 95% confidence interval（CI）（horizontal line）；pooled evidence（large black diamond）；relative risk of 1.0（vertical black line）.
Abbreviations：MIDHS，minimally invasive dynamic hip screw；CDHS：conventional dynamic hip screw；NA，not available；M/F，male/female；RCT was for randomized controlled trial
References：
1. Freeman C，Todd C，Camilleri-Ferrante C，Laxton C，Murrell P，Palmer CR，et al.，Quality improvement for patients with hip fracture：experience from a multi-site audit. Qual Saf Health Care，2002. 11（3）：p. 239-45.
2. Roberts SE and Goldacre MJ，Time trends and demography of mortality after fractured neck of femur in an English population，1968-98：database study. BMJ，2003. 327（7418）：p. 771-5. 3. Cole PA and Bhandari M，What's new in orthopaedic trauma. J Bone Joint Surg Am，2005. 87（12）：p. 2823-38.
4. Alobaid A，Harvey EJ，Elder GM，Lander P，Guy P，and Reindl R，Minimally invasive dynamic hip screw：prospective randomized trial of two techniques of insertion of a standard dynamic fixation device. J Orthop Trauma，2004. 18（4）：p. 207-12.
5. Shams A，El-Sayed M，Elsawy M，Hafez K，and Gad H，Comparative，prospective，randomized study of the modified minimally invasive technique versus the conventional technique of dynamic hip screw（DHS），fixation for intertrochanteric fractures in the elderly. European Orthopaedics and Traumatology，2015. 6（1）：p. 27-33.
6. Wang JP，Yang TF，Kong QQ，Liu SJ，Xiao H，Liu Y，et al.，Minimally invasive technique versus conventional technique of dynamic hip screws for intertrochanteric femoral fractures. Arch Orthop Trauma Surg，2010. 130（5）：p. 613-20.
7. Zhou Z，Zhang X，Tian S，and Wu Y，Minimally invasive versus conventional dynamic hip screw for the treatment of intertrochanteric fractures in older patients. Orthopedics，2012. 35（2）：p. e244-9.
8. Moher D，Liberati A，Tetzlaff J，and Altman DG，Preferred reporting items for systematic reviews and meta-analyses：the PRISMA statement. Int J Surg，2010. 8（5）：p. 336-41.
9. Lee YS，Huang HL，Lo TY，and Huang CR，Dynamic hip screw in the treatment of intertrochanteric fractures：a comparison of two fixation methods. Int Orthop，2007. 31（5）：p. 683-8.
10. Wang XL，Qin J，Kuang JX，Chen M，Ge T，et al.，Comparative study of internal fixation with minimally invasive and traditional dynamic hip screw in the treatment of intertrochanteric fractures in aged people. Anatomy and C linics，2007（04）：p. 264-267.
11. Wong TC，Chiu Y，Tsang WL，Leung WY，and Yeung SH，A double-blind，prospective，randomised，controlled clinical trial of minimally invasive dynamic hip screw fixation of intertrochanteric fractures. Injury，2009. 40（4）：p. 422-7.
12. Zhong HH，Clinical comparison of minimally invasixe and conventional dynamic hip screw in the treatment of femoral intertrochanteric fractures. China Journal of Modern Medicine，2011（34）：p. 4335-4336+4340.
13. Chen Z，Zhou B，Minimally invasive dynamic hip screw in treatment of intertrochanteric fracture：A report of 30 cases. China Modern Doctor，2012. 50（33）：p. 126-127.
14. Bonnaire F，Zenker H，Lill C，Weber AT，and Linke B，Treatment strategies for proximal femur fractures in osteoporotic patients. Osteoporos Int，2005. 16 Suppl 2：p. S93-S102.
15. Perkins HA，Transfusion-induced immunologic unresponsiveness. Transfus Med Rev，1988. 2（4）：p. 196-203.
16. Dodd RY，The risk of transfusion-transmitted infection. N Engl J Med，1992. 327（6）：p. 419-21.
17. Linden JV and Kaplan HS，Transfusion errors：causes and effects. Transfus Med Rev，1994. 8（3）：p. 169-83.
18. Hoeksma HL，Van Den Ende CH，Ronday HK，Heering A，and Breedveld FC，Comparison of the responsiveness of the Harris Hip Score with generic measures for hip function in osteoarthritis of the hip. Ann Rheum Dis，2003. 62（10）：p. 935-8.
19. Kunz R，Vist G，and Oxman AD，Randomisation to protect against selection bias in healthcare trials. Cochrane Database Syst Rev，2007（2）：p. MR000012.
20. Simmonds MC，Higgins JP，Stewart LA，Tierney JF，Clarke MJ，and Thompson SG，Meta-analysis of individual patient data from randomized trials：a review of methods used in practice. Clin Trials，2005. 2（3）：p. 209-17.
Method We searched the database of PubMed，EMBASE，the Cochrane Library，CNKI and VIP from inception to August 2015 for relevant randomized controlled trials（RCTs）without language restriction. References of all the selected articles were hand-searched for any additional trials. Two authors independently extracted data from all eligible studies，including study design，participants，interventions，and outcomes（wound size，surgery time，intraoperative blood loss，length of hospital stay，Harris hip score and severe complications）. Data were analyzed using fixed-effects and random-effects models with mean differences and risk ratios for continuous and dichotomous variables，respectively.
Results Eight studies involving 612 patients were identified in this analysis. The meta-analysis showed smaller wound size（WMD，-7.12；95% CI，-7.68，-6.56，respectively），shorter surgery time（WMD，-21.86；95% CI，-36.09，-7.62，respectively），less intraoperative blood loss（WMD = 4.60 mL，95%CI，227.06–36.27 ml），shorter length of hospital stay（WMD，-3.30；95% CI，-3.72，-2.88，respectively），higher Harris hip score（WMD，1.54；95% CI，0.40，2.67，respectively），and fewer severe complications（RR，0.47，95 % CI 0.22，1.00）in patients receiving MIDHS versus those receiving CDHS.
Conclusions The available evidence suggests that MIDHS is superior to CDHS for intertrochanteric fractures in elderly patients.
Introduction
Hip fractures are becoming increasingly common with the aging of the population and approximately half of these fractures occur in the intertrochanteric region. These fractures typically occur in elderly patients and often result in eventual loss of their functional independence[1，2]. Intertrochanteric fractures in the elderly are associated with higher rates of mortality and complications if they fail to mobilize or ambulate early. For a return to pre-injury functioning and activity levels，early operative interventions are recommended for senile femoral intertrochanteric fracture[3]. The dynamic hip screw has become the standard and is currently the most common implant for intertrochanteric fractures[4]. Several prospective randomized controlled trials（RCT）over the previous decade have shown that minimally invasive dynamic hip screw（MIDHS）can reduce intraoperative blood loss，shorten operative time，and relieve postoperative pain[4-6]. No well-designed meta-analysis and systematic review capable of providing high levels of evidence has been conducted. Recently，a systematic review and meta-analysis were conducted to test the theoretical advantages of MIDHS fixation over CDHS fixation for intertrochanteric hip fractures[7]. However the paper also included quasi- and non-randomized controlled trials（RCT），which reduces the confidence level of the meta-analysis. Therefore，we carried out this meta-analysis to evaluate and compare the clinical outcomes and safety of MIDHS vs. CDHS from published RCTs.
Materials and methods
Search Strategies
We searched the database of PubMed，EMBASE，Cochrane Library，CNKI and VIP for all articles on MIDHS and CDHS for intertrochanteric hip fracture without language restriction. The following search terms were used：（‘intertrochanteric hip fracture’ OR ‘fractures’）AND（randomized controlled trial）. The search was performed in August，2015 and covered literatures published between inception and August，2015. We also retrieved relevant articles in Google Scholar. The bibliographies of the obtained literature and the references of relevant meta-analyses were checked to ensure that no related studies were inadvertently neglected. This meta-analysis was performed in accordance with the PRISMA guidelines[8].
Selection criteria
The inclusion criteria were as follows：1）patients had intertrochanteric fractures，and 2）studies compared MIDHS with CDHS. When two studies were reported by the same institution and authors，the study with the higher quality or the most recent publication was included in the analysis，unless the study outcomes were mutually exclusive or measured at different time points. Exclusion criteria were as follows：1）data collection was not scientific and data analysis was incorrect or not provided，and 2）review literature，repeated reports and retrospective studies.
Data extraction
All data were extracted independently by two authors according to the inclusion criteria. Disagreements were resolved by discussion between the two reviewers. The following characteristics were collected from each study：the first author，year of publication，source，experimental design，sample size，sample characteristics，and primary and secondary outcomes. Un-weighted κ statistics was applied to assess agreement between the two reviewers. The quality of the included studies was assessed using the Newcastle-Ottawa Scale.
Statistical analysis Statistical analysis was conducted using Review Manager 5.2 software. Continuous data were expressed as the weighted mean difference（WMD）and 95 % confidence intervals（CI）. Dichotomous data were presented as risk ratios（RR）with 95 % CI. P ≤ 0.05 was considered statistically significant. Heterogeneity was tested using the chi-square test and the I2 test. A significance level of 0.1 for the chi-square test was interpreted as evidence of heterogeneity. The I2 test was used to estimate the total variation among the studies. When no statistical evidence of heterogeneity was present，a fixed effects model was adopted. If not，a random effects model was chosen.
Results
Characteristics of the included studies
The study flowchart is shown in Fig. 1. Three hundred and twelve publications were identified for full text assessment with the search strategy（91 from PubMed，33 from the Cochrane Library，79 from Embase，42 from CNKI，62 from WanFang and 5 from the references of previous meta analyses）. After exclusion of duplications using Notexpress software and manual confirmation，214 publications without duplications remained. The two reviewers further read the articles and finally 8 papers were included[4-6，9-13].The quality of the included studies was using the Newcastle-Ottawa Scale for all 8 studies. There was excellent agreement between investigators for full text screening（κ= 0.92）. The basic characteristics of these studies such as the authors，publication year，journal，study type，the number of patients，treatment method and follow-up duration are summarized in Table 1. The number of patients in each study ranged from 53 to132 and the total number of patients was 612，including 302 patients receiving MIDHS and 310 patients receiving CDHS. The methodological quality of included studies is presented in Fig. 2.
Wound size
Five RCTs[6，9，10，12，13]studied wound size after MIDHS and CDHS for intertrochanteric fractures. A total of 430 patients were studied，including 204 patients receiving MIDHS and 226 patients receiving CDHS. No heterogeneity was present among the studies（P=0.07，I2=53 %）. There was a statistically significant difference in wound size between MIDHS and CDHS for intertrochanteric fractures（WMD，-7.12；95% CI，-7.68，-6.56，respectively），indicating that MIDHS can significantly reduce wound size（Fig. 3）.
Operative time
Five RCTs [6，9-12]reported the duration of surgery using MIDHS or CDHS for intertrochanteric fractures. MIDHS was used for 209 patients and 227 patients for CDHS. Heterogeneity tests indicated statistical evidence of heterogeneity（P<0.00001，I2=96 %）. Therefore，a random effects model was used. The meta-analysis showed a significant difference in operative time between MIDHS and CDHS for intertrochanteric fractures（WMD，-21.86；95% CI，-36.09，-7.62），indicating MIDHS treatment requires less surgery time（Fig. 4）. Intraoperative blood loss
Four studies[6，10，12，13]reported data on intraoperative blood loss. Due to high heterogeneity（P < 0.00001，I2 = 92%），a random-effects model was adopted to pool the data. There was a significant difference between the two groups regarding intraoperative blood loss（WMD = -202.99 ml，95% CI，-221.57，–81.80 mL；P = 0.25）（Fig. 5）. A further sensitivity analysis，which was performed after one RCT15was excluded，indicated that no significant difference（WMD = 4.60 ml，95%CI，227.06–36.27 mL；P = 0.78；fixed-effects model）with low heterogeneity（P = 0.31，I2 = 14%）between the two groups（Fig. 5）.
Length of hospital stay
Three RCTs[9，11，12] reported the length of hospital stay after MIDHS and CDHS for intertrochanteric fractures. One hundred forty-nine patients were included in the MIDHS group and 161 patients in the CDHS group. No heterogeneity was observed between the studies（P= 0.64，I2 = 0 %）. Therefore，a fixed effects model was used. Our meta-analysis showed a statistical difference in the length of hospital stay between MIDHS and CDHS for intertrochanteric fractures（WMD，-3.30；95% CI，-3.72，-2.88）. MIDHS significantly reduced the length of hospital stay（Fig. 6）.
Harris hip score
Five RCTs [5，6，9，11，13] reported Harris hip scores after MIDHS and CDHS treatment of intertrochanteric fractures. One hundred forty-nine patients were included in the MIDHS group and 16 patients in the CDHS group. No heterogeneity was present between the studies（P = 0.64，I2 = 0 %）.Therefore，a fixed effects model was used. The meta-analysis demonstrated a statistically significant difference in Harris hip scores between MIDHS and CDHS for intertrochanteric fractures（WMD，1.54；95% CI，0.40，2.67）. The postoperative Harris hip score was significantly higher in the MDHS group（Fig. 7）.
Severe complications
Six RCTs[4，6，9-11，13]provided data on severe complication rate after MIDHS and CDHS treatment for intertrochanteric fractures. A total of 447 patients were studied，including 213 patients receiving MIDHS treatment and 234 patients receiving CDHS treatment. No heterogeneity was present between the studies（P=0.05，I2 =3 %）. The result of meta-analysis showed a statistical difference in the neurological complication rate between the MIDHS group and the CDHS group（RR 0.47，95 %）（Fig. 8）. In the CDHS group，6 patients developed deep vein thrombosis（DVT）and 2 patients had pulmonary embolism，6 patients had implant failure，1 patient developed atrial fibrillation，and 1 patient had urinary tract infection. In addition，5 patients died. In the MIDHS group，4 patients developed DVT，and 2 paitents had implant failure. Two patients died in the group. Discussion
The treatment of intertrochanteric femoral fractures remains a challenge for orthopedic surgeons due to lack of unified standard in fracture classification，treatment，and curative effect evaluation. Minimally invasive surgery has gained in popularity in orthopedic surgery as it is associated with reduced operation time，smaller blood loss and decreased postoperative pain，reduced postoperative morbidity，and faster recovery of function. The purpose of this meta-analysis and systematic review was to summarize the currently available evidence to determine the safety and efficacy of the minimally invasive technique，MIDHS，vs. CDHS for intertrochanteric femoral fractures in the elderly. This meta-analysis indicates that MIDHS and CDHS differ significantly in wound size，operative time，intraoperative blood loss，length of hospital stay，Harris hip score，and severe complications.
Intertrochanteric femoral fractures in the elderly are associated with higher rates of co-morbidities such as DVT，urinary tract infections，and pulmonary embolism[14]. A reduced operative time，especially in elderly patients with comorbid conditions or poor cardiopulmonary reserve，is desirable because it reduces temporal exposure to the risks of general anesthesia. This，combined with reduced surgical trauma，may be significant in reducing postoperative morbidity and mortality in such patients. Intraoperative blood loss in MIDHS was reduced significantly compared with CDHS. This can be attributed to smaller incisions，less soft-tissue damage，and minimally operative trauma without exposing the fracture in MIDHS. Smaller intraoperative blood loss reduces the need for and avoids the hazards of blood transfusion for elderly patients[15-17]，thus helping minimize the healthcare cost for intertrochanteric hip fracture，which imposes a mounting economic burden on the patients，the family and the society.
Early mobilization largely depends on the reduction of postoperative pain，and some studies show a trend toward lower pain scores in patients receiving MIDHS [4，11].which may encourage earlier ambulation and，theoretically，facilitate earlier discharge. MIDHS requires less operative time，reduces the amount of intraoperative blood loss，and allows patients to be discharged sooner，which may benefit patients’ recoveries and has significant financial implications for hospitals. The rate of severe postoperative complications in the MIDHS group（3.8%）was markedly lower than that of the CDHS group（8.5%）. Elderly patients are at higher risks for DVT，urinary tract infections，and pulmonary embolism if they fail to mobilize or ambulate early. Reduced operative time，less bleeding，and less postoperative pain may promote earlier ambulation in patients receiving MIDHS，which may contribute to a more favorable outcome in terms of severe postoperative complications. The implant failure rates are similar between patients receiving MIDHS and those receiving CDHS，possibly because they are unstable fractures with loss of medial calcar continuity and with osteoporosis，and bone grafting is not done intraoperatively[6，10]. The MIDHS is safer and more dependable than the CDHS because of severe postoperative complications and should be the first option for treatment of intertrochanteric fractures. Harris hip score is a multidimensional observational assessment of pain，walking function，activities of daily living，and hip joint range of motion[18]. Our meta-analysis showed significantly higher scores of patients receiving MIDHS than those receiving CHDS，implying a faster functional recovery of patients receiving MIDHS. Because intertrochanteric fractures usually occur in elderly patients who may have age-associated cognitive deficits or other diseases，it is important for elderly patients to return to pre-injury activity levels as soon as possible to avoid complications. Higher Harris hip scores of the MIDHS group suggest MIDHS may be more beneficial for elderly patients.
Our study focus on systematic reviews of RCTs，which are likely to provide more reliable information than other sources of evidence on the differential effects of alternative forms of healthcare[19]. Undoubtedly，this analysis also has some limitations. Significant heterogeneity was observed between the included trials for intraoperative blood loss，and operative time. This heterogeneity may be attributable to variation in the skills of the surgeons and the different types of peritrochanteric fractures. The eligibility criteria for inclusion of patients with peritrochanteric fractures differ from one another，which might influence consistency of effects across the included studies and cause between-study heterogeneity. To ensure uniformity in both defining patients’ characteristics for peritrochanteric fractures and defining outcome measures，an individual patient data meta-analysis is needed[20]. Moreover，we did not undertake a subgroup analysis of the different fracture types to identify the possible source of heterogeneity because not all of the included studies described data according to different fracture types. Furthermore，the effects might differ by different fracture types，which require further study.
Conclusions
Based on this systematic review and meta-analysis，MIDHS is superior to CHDS in terms of wound size，surgery time，intraoperative blood loss，length of hospital stay，Harris hip score and severe complications. Thus，the MIDHS technique is recommended as a minimally invasive technique that can be considered as an additional alternative treatment for intertrochanteric fractures，especially in elderly patients with multiple morbidities.
Conflict of interest
The authors declare that they have no conflict of interest.
Figure legends Figure 1 The study flowchart.
Figure 2 Methodological quality of included studies. The risk of bias tool incorporates assessment of randomization（sequence generation and allocation concealment），blinding（participants，personnel，and outcome assessors），completeness of outcome data，selection of outcomes reported，and other sources of bias. The items are scored with “yes，” “no，” and “unclear.”
Figure 3 Comparison of wound size（cm）between the MIDHS and CDHS group. Key relative risk（square）and its 95% confidence interval（CI）（horizontal line）；pooled evidence（large black diamond）；relative risk of 1.0（vertical black line）.
Figure 4 Comparison of operative time（minutes）between the MIDHS and CDHS treatments. Key relative risk（square）and its 95% confidence interval（CI）（horizontal line）；pooled evidence（large black diamond）；relative risk of 1.0（vertical black line）.
Figure 5 Comparison of intraoperative blood loss（mL）between the MIDHS and CDHS group. Key relative risk（square）and its 95% confidence interval（CI）（horizontal line）；pooled evidence（large black diamond）；relative risk of 1.0（vertical black line）.
Figure 6 Comparison of length of hospital stay（days）between the MIDHS and CDHS group. Key relative risk（square）and its 95% confidence interval（CI）（horizontal line）；pooled evidence（large black diamond）；relative risk of 1.0（vertical black line）.
Figure 7 Comparison of postoperative Harris hip score between the MIDHS and CDHS group. Key relative risk（square）and its 95% confidence interval（CI）（horizontal line）；pooled evidence（large black diamond）；relative risk of 1.0（vertical black line）.
Figure 8 Comparison of serious postoperative complication rates between the MIDHS and CDHS treatments. Key relative risk（square）and its 95% confidence interval（CI）（horizontal line）；pooled evidence（large black diamond）；relative risk of 1.0（vertical black line）.
Abbreviations：MIDHS，minimally invasive dynamic hip screw；CDHS：conventional dynamic hip screw；NA，not available；M/F，male/female；RCT was for randomized controlled trial
References：
1. Freeman C，Todd C，Camilleri-Ferrante C，Laxton C，Murrell P，Palmer CR，et al.，Quality improvement for patients with hip fracture：experience from a multi-site audit. Qual Saf Health Care，2002. 11（3）：p. 239-45.
2. Roberts SE and Goldacre MJ，Time trends and demography of mortality after fractured neck of femur in an English population，1968-98：database study. BMJ，2003. 327（7418）：p. 771-5. 3. Cole PA and Bhandari M，What's new in orthopaedic trauma. J Bone Joint Surg Am，2005. 87（12）：p. 2823-38.
4. Alobaid A，Harvey EJ，Elder GM，Lander P，Guy P，and Reindl R，Minimally invasive dynamic hip screw：prospective randomized trial of two techniques of insertion of a standard dynamic fixation device. J Orthop Trauma，2004. 18（4）：p. 207-12.
5. Shams A，El-Sayed M，Elsawy M，Hafez K，and Gad H，Comparative，prospective，randomized study of the modified minimally invasive technique versus the conventional technique of dynamic hip screw（DHS），fixation for intertrochanteric fractures in the elderly. European Orthopaedics and Traumatology，2015. 6（1）：p. 27-33.
6. Wang JP，Yang TF，Kong QQ，Liu SJ，Xiao H，Liu Y，et al.，Minimally invasive technique versus conventional technique of dynamic hip screws for intertrochanteric femoral fractures. Arch Orthop Trauma Surg，2010. 130（5）：p. 613-20.
7. Zhou Z，Zhang X，Tian S，and Wu Y，Minimally invasive versus conventional dynamic hip screw for the treatment of intertrochanteric fractures in older patients. Orthopedics，2012. 35（2）：p. e244-9.
8. Moher D，Liberati A，Tetzlaff J，and Altman DG，Preferred reporting items for systematic reviews and meta-analyses：the PRISMA statement. Int J Surg，2010. 8（5）：p. 336-41.
9. Lee YS，Huang HL，Lo TY，and Huang CR，Dynamic hip screw in the treatment of intertrochanteric fractures：a comparison of two fixation methods. Int Orthop，2007. 31（5）：p. 683-8.
10. Wang XL，Qin J，Kuang JX，Chen M，Ge T，et al.，Comparative study of internal fixation with minimally invasive and traditional dynamic hip screw in the treatment of intertrochanteric fractures in aged people. Anatomy and C linics，2007（04）：p. 264-267.
11. Wong TC，Chiu Y，Tsang WL，Leung WY，and Yeung SH，A double-blind，prospective，randomised，controlled clinical trial of minimally invasive dynamic hip screw fixation of intertrochanteric fractures. Injury，2009. 40（4）：p. 422-7.
12. Zhong HH，Clinical comparison of minimally invasixe and conventional dynamic hip screw in the treatment of femoral intertrochanteric fractures. China Journal of Modern Medicine，2011（34）：p. 4335-4336+4340.
13. Chen Z，Zhou B，Minimally invasive dynamic hip screw in treatment of intertrochanteric fracture：A report of 30 cases. China Modern Doctor，2012. 50（33）：p. 126-127.
14. Bonnaire F，Zenker H，Lill C，Weber AT，and Linke B，Treatment strategies for proximal femur fractures in osteoporotic patients. Osteoporos Int，2005. 16 Suppl 2：p. S93-S102.
15. Perkins HA，Transfusion-induced immunologic unresponsiveness. Transfus Med Rev，1988. 2（4）：p. 196-203.
16. Dodd RY，The risk of transfusion-transmitted infection. N Engl J Med，1992. 327（6）：p. 419-21.
17. Linden JV and Kaplan HS，Transfusion errors：causes and effects. Transfus Med Rev，1994. 8（3）：p. 169-83.
18. Hoeksma HL，Van Den Ende CH，Ronday HK，Heering A，and Breedveld FC，Comparison of the responsiveness of the Harris Hip Score with generic measures for hip function in osteoarthritis of the hip. Ann Rheum Dis，2003. 62（10）：p. 935-8.
19. Kunz R，Vist G，and Oxman AD，Randomisation to protect against selection bias in healthcare trials. Cochrane Database Syst Rev，2007（2）：p. MR000012.
20. Simmonds MC，Higgins JP，Stewart LA，Tierney JF，Clarke MJ，and Thompson SG，Meta-analysis of individual patient data from randomized trials：a review of methods used in practice. Clin Trials，2005. 2（3）：p. 209-17.