Operative Treatment of HIP Fractures in Patients on Clopidogrel: A Case-Control Study

Abstract

Introduction

Clopidogrel, an inhibitor of ADP-induced platelet aggregation, is indicated for the reduction of atherosclerotic events in patients with atherosclerosis documented by recent stoke, myocardial infarction, acute coronary syndrome, and established peripheral arterial disease. In cardiovascular studies, clopidogrel has been associated with increased chest tube output, transfusion rates, and re-exploration rates. Few studies have addressed the possible complications of clopidogrel in hip fractures. Our study aims to assess the perioperative blood loss and transfusion rates in geriatric patients with hip fractures on clopidogrel. We hypothesize that patients on clopidogrel will have higher perioperative blood loss and transfusion rates.

Materials and Methods

A retrospective, case control study chart review over a five year span was conducted. Of the 2,766 geriatric hip fracture patients surgically treated, 52 patients taking clopidogrel upon admission to the hospital were compared to patients not on the drug. All of the patients in the study were taken to the operating room within two calendar days of admission. statistical analysis was performed using Wilcoxon's, Fisher exact, chi square, and logistic regression methods.

Results

A total of 110 patients were included in the analysis, 52 (47%) were taking clopidogrel at the time of admission. these patients were compared to 58 (53%) patients not on the drug. No significant difference was found with respect to documented perioperative blood loss. Transfusion rates however, did vary. Patients who had been taking clopidogrel, prior to admission and subsequent surgery, had a transfusion rate of 56% while those patients not on the drug had a transfusion rate of 31%. Logistic regression analysis showed taking clopidogrel up to admission was significantly associated (p = .0121) with receiving a blood transfusion following surgical treatment of a hip fracture.

Conclusion

A growing body of evidence supports early (within 48 hrs) surgery for elderly patients with hip fractures. the pharmacokinetics of clopidogrel do not allow for bleeding time to return to normal until the drug has been discontinued for five days. Our study shows that patients taking clopidogrel upon admission for hip fracture are at increased risk of blood transfusions when surgery is performed within two calendar days of admis-sion. this risk must be balanced by the potential benefits of early surgery.

Introduction

Clopidogrel is an irreversible inhibitor of platelet aggregation and thereby increases bleeding time. Clopidogrel blocks adenosine diphosphate (ADP) from binding to the platelet receptor (P2Y12) and the subsequent ADP-mediated activation of the glycoprotein GPIIb/IIIa complex.^{1, 2} Clopidogrel has a half life of eight hours and renders platelets inactive for approximately five to seven days, the average life span of a platelet. Platelet aggregation and bleeding time gradually return to base-line values five days after treatment is discontinued¹ Clopidogrel is indicated for the reduction of atherosclerotic events in patients with atherosclerosis documented by recent stoke, myocardial infarction, acute coronary syndrome, and established peripheral arterial disease. Furthermore, clopidogrel is now widely used postoperatively in patients undergoing cardiac and carotid stenting. A search of our institution's database revealed that the number of patients taking clopidogrel has more than doubled in the past five years.

Clopidogrel and its effects on surgical patients have been widely studied in the cardiac literature. Yende and Wunderink³ concluded that clopidogrel in combination with aspirin is associated with an increased need for surgical re-exploration as well as risk of red blood cell and cryoprecipitate transfusions following coronary artery bypass surgery. Furthermore, Von Heymann et al⁴ reported that aspirin and clopidogrel taken until two days prior to coronary artery bypass grafting is associated with significantly higher chest tube outputs. Their study also confirmed previous reports regarding higher rates of re-exploration secondary to bleeding in patients on aspirin and clopidogrel. With the increasing number of patients taking clopidogrel, studies continue to examine the implications of the irreversible platelet inhibitor in the surgical setting.

In orthopaedics, hip fracture patients comprise a particular group that may be impacted by the use of clopidogrel. The number of patients with hip fractures is increasing. The World Health Organization estimates that by 2050, a total of 6 million fractures of the hip will occur worldwide every year.⁵ The combination of increasing population age and longer life spans, the increasing use of clopidogrel, and the increasing number of hip fractures creates a unique issue and potential concern for orthopaedic surgeons. Furthermore, hip fractures are a major source of morbidity and mortality in the elderly population. Mortality in the Medicare population for those who sustain a hip fracture is estimated to be 13% at three months and 24% at one year.⁶ Multiple studies have shown the benefit of early surgery (within 48 hours). Reported benefits include decreased mortality and shorter hospital stays.^{7, 8} With the goal of early surgical intervention to reduce mortality, hip fracture patients on clopidogrel have dysfunctional platelets and potentially increased bleeding times secondary to the pharmacokinetics of clopidogrel when surgery is performed. Simple bleeding time and platelet function have not returned to baseline within two days of stopping the drug.

The aim of this study is to compare blood loss and transfusion rates in elderly hip fracture patients on clopidogrel who were treated operatively within two calendar days of admission versus those who were not on the irreversible platelet aggregation inhibitor. By doing so we hoped to potentially provide additional insight in the complex management of geriatric hip fracture patients.

Materials and Methods

An institutional review board approval was obtained prior to our study. All patients were treated at an academic level one trauma center. A retrospective case control study was performed in order to investigate blood loss and transfusion requirements of patients who underwent open treatment of a hip fracture, including femoral neck and intertrochanteric fractures over a five year time frame from 2000 - 2005. Exclusion criteria included 1) use of another anticoagulant or blood thinner except aspirin in combination with clopidogrel at the time of admission; 2) presence of any blood dyscrasia or hemoglobinopathy; and 3) delay of surgical intervention greater than two days. All patients were evaluated by an inpatient internal or family medicine team prior to operative treatment. All patients were medically optimized prior to surgical intervention. Clopidogrel was held prior to surgery in all cases and in a majority not resumed until two weeks post operative.

Patients taking clopidogrel upon admission/fracture were compared to a control group with no clopidogrel use. Using our electronic medical record, medical record numbers were obtained meeting the criteria of hip fracture. The search provided 2,766 patients who underwent surgical management of a hip fracture. Eighty-five of these patients were taking clopidogrel at the time of admission. Fifty-two patients met inclusion and exclusion criteria. Data collected for each patient included age, sex, time to surgery, method of surgical treatment, comorbidities, amount of intraoperative fluids, post operative transfusion, quantity of units transfused, intra-operative blood loss, and preoperative and postoperative hemoglobin levels. A total of 58 control subjects were matched to cases with regard to sex, age, comorbidities and surgical procedure performed. All members of the case group were taking clopidogrel at 75 mg per day at the time of admission. Patients in the control group had either not been on clopidogrel or had discontinued the drug at least one week prior to admission.

The amount of blood loss was recorded from the dictated operative report and anesthesia records from surgery. Transfusion requirements were noted in the medical record and dictated discharge summary. The decision to transfuse was at the discretion of the surgeon with consideration of the patient’s vital signs, medical history, laboratory values and recommendations from our medicine colleagues. The Biostatistics department helped with data analysis. A p-value of 0.05 or less with a confidence interval of 95% was considered significant. Wilcoxon's, Fisher exact, chi square, and logistic regression methods were utilized. We hypothesized that patients on clopidogrel upon admission for a hip fracture would have both increased blood loss and transfusion rates.

Results

A total of 110 patients were included in the study. The clopidogrel group consisted of 52 patients (14 males and 38 females). The control group consisted of 58 patients (16 males and 42 females). There was no statistical significance in gender between the two groups. The average age was 82.4 for the clopidogrel group and 77.6 in the control group. This was a statistical difference (p value = 0.009). Table 1 summarizes preoperative characteristics between the clopidogrel group and the control group (Table 1). With regards to comorbidities, the only statistical difference was found in history of stroke and transient ischemic attack between the clopidogrel and control groups (25 vs 11 respectively). Table 2 summarizes the medical comorbidities documented upon admission. No significant difference existed in the incidence of preoperative transfusion between the two groups (Table 2). The preoperative hemoglobin was 11.8 gm/dl +/- 1.5 in the group taking clopidogrel versus 12 gm/dl +/- 2 in the control group

Table 1.

Preoperative Characteristics Preoperative demographics of age, gender, and transfusions of patients on clopidogrel and patients not on clopidogrel. Only age was statistically significant between the two groups.

Variable	Clopidgrel Control (N=52)	Clopidgrel Control (N=58)	p-value
Gender
Male	14	16	0.937
Female	38	42
Age	82.4 +/- 8.9	77.6 +/- 8.7	0.00911
Pre-operatie Transfusion
Yes	9	7	0.55922
No	43	46

Chi-Square test was used unless otherwise stated.

¹Wilcoxon's test was used

²Fisher's exact test was used

Table 2.

Medical Comorbidities Distribution of medical comorbidities between patients on clopidogrel and patients not on clopidogrel. History of stroke/TIA was statistically significant between the two groups.

Comorbidities	Clopidogrel Group (N=52)	Control Group (N=58)	p-value
Cardiovascular Disease/Hypertension	50	53	.044311
Stroke/TIA	25	11	0.001
COPD	7	12	0.316
Renal Disease	3	10	0.063
Diabetes	13	23	0.102
Malignancy	1	4	0.3671
Current Smoker	8	8	0.813
Other	3	10	0.063

Chi-Square test was used unless specified

¹Fisher's exact test was used

Table 3 details the procedures performed on the 110 patients included in the study. No significant difference was found between the two groups among method of surgical treatment (Table 3). Selection of the procedure was at the discretion of the treating surgeon based on the patient's fracture pattern and location. Table 3 also shows the recorded blood loss during surgery. Blood loss was averaged among the two groups as a whole and not studied specific to each of the four procedures performed. No significant difference was found between the two groups in regards to the procedure performed or intraoperative blood loss. There was also no significant difference in surgery time between the two groups with the study group averaging 65 +/- 40 minutes versus the control group averaging 59 +/- 38 minutes.

Table 3.

Operative Procedure/Blood Loss Distribution of surgical procedure and intraoperative blood loss between patients on clopidogrel and patients not on clopidogrel.

Procedure	Clopidogrel Group (N=52)	Control Group (N=58)
Gamma Nail	17	18
Endoprosthesis	5	7
Dynamic Hip Screw	22	22
Percutaneous	8	11
Pinning
Blood Loss	142ml +/- 138	137ml +/- 120

Chi -squared test was used

The lowest documented hemoglobin was 9 mg/dl +/1.9 in the clopidogrel group versus 9.5 mg/dl +/- 1.7 in the control group, which was not significant (p = 0.41). There was, however, a significant difference in the transfusion rate between the two groups of patients (see figure 1). Twenty-nine of 52 (56%) patients taking clopidogrel were transfused postoperatively versus 18 of 58 (31%) patients in the control group (p = 0.008). Since age and transfusion rates were significantly different among the two groups, a logistic regression analysis was performed. The results, (Table 4) demonstrate that the clopidogrel group had a statistically higher risk of transfusion when compared to patients not on clopidogrel (p = 0.012). In our patients, age did not increase the probability of a postoperative transfusion.

Figure 1. Transfusion rate.

Transfusion rate between patients on clopidogrel and patients not on clopidogrel.

Table 4.

Risk Factors for Transfusion Analysis of variables for transfusion.

Variable	p-value
Clopidogrel versus control group	0.012
Age	0.965

Logistic Regression Analysis: Categorical variable considered were are and consumption of clopidogrel. The response variable is transfusion postoperatively.

Discussion and Conclusion

Multiple studies support early operative intervention for patients with hip fractures.^8-10 Zuckerman et al⁹ prospectively illustrated operative delay beyond two calendar days doubled the risk of death within the first postoperative year. Gdalevich¹⁰ and McGuire⁸ found an increased mortality rate associated with surgical delay. Several studies have demonstrated increased mortality rate associated with surgical delay.^8,10-12 Length of hospital stay has also shown to be dependent on surgical delay. Siegmuth et al⁷ found a significant increase in length of stay in patients operated on after 48 hours. Furthermore, there are economic consequences associated with operative delay in the treatment of hip fractures. Shabat¹³ recommended surgery within the first 48 hours, with early surgery being more cost effective with shorter hospital stays. Therefore, with goal of early intervention, delaying treatment until platelet function has returned is not ideal.

Recently, there have been two studies that examined the impact of clopidogrel in orthopaedic patients. Nydick¹⁴ illustrated no increase in complications or transfusion requirements in patients undergoing non-elective orthopaedic surgery with clopidogrel. The clopidogrel group was taken to surgery within five days whereas our study looked at surgical intervention within two days. At two days the platelet function is still quite inhibited and our study demonstrates the effect of clopidogrel in early surgical treatment for hip fractures. Christy¹⁵ demonstrated no difference in early (<24 hours) or late (>24 hours) transfusion rates in patients on clopidogrel with pelvic fractures.

Our study examined the impact clopidogrel had on blood loss and transfusion rates in early (less than two days) operative treatment of elderly hip fractures. There was no statistical difference in intra-operative blood loss or preoperative transfusion rates; however, there was a statistical difference in postoperative transfusion rates (p = 0.012), age (p = 0.009), and history of stroke/transient ischemic attack (TIA) (p = 0.001) in the clopidogrel group. On regression analysis, age did not appear to be a significant factor to transfusion rates.

There are several areas of improvement to our study. Our study is a retrospective analysis with a relatively small number of patients. Transfusion risk with respect to specific procedures was not analyzed; rather, all four procedures were studied as a whole. We were unable to control for bias with respect to transfusion rate. There was no set criterion for decision to transfuse; however, no publications to date have demonstrated transfusion guidelines in management of hip fracture patients. A majority of transfusions are individualized, multifactorial, and based on experience of the treating physician. A prospective study with uniform transfusion criteria may further clarify the impact of clopidogrel. Furthermore, a higher percentage of patients on clopidogrel had a positive past medical history for stroke or TIA. With lack of a uniformly applied transfusion protocol, physicians treating these patients may have had lower thresholds for transfusion compared to patients without a history of stroke or TIA. This is a significant weakness of the study. Lastly, the strength of the study would have been improved by expanding our analysis to include patients taking aspirin.

In this study hip fracture patients on clopidogrel at time of admission had an increased risk of receiving a blood transfusion after early operative intervention. No increase in perioperative morbidity could be otherwise linked to the use of clopidogrel. Counseling patients on clopidogrel regarding the risk of transfusion for operative treatment of hip fractures should be considered; however, the benefits of early surgical intervention likely outweigh the risks of transfusion in patients sustaining hip fractures while on clopidogrel.