Hepatitis C is a worldwide health problem, affecting 1.8% of the population in the United States and >3% of the world population1. In 1980, Alter et al. noted that most post-transfusion hepatitis cases were due to a non-A, non-B virus2,3. This virus was subsequently named the hepatitis-C virus in 1989. Acute hepatitis C is a parenterally transmitted viral infection, often presenting with only mild symptoms within the first eight weeks after exposure. Progression to chronic hepatitis occurs in 74% to 86% of patients, with a broad range of clinical manifestations, from asymptomatic viremia to cirrhosis, progressive liver failure, and hepatocellular carcinoma4.
As the interval between infection and the development of cirrhosis may exceed thirty years, many chronically infected, non-cirrhotic patients seek treatment for degenerative joint disease4. Although some studies have suggested increased morbidity and mortality in patients with liver cirrhosis who undergo surgery5,6, little is known about the effects of hepatitis-C infection on the outcome of total hip and knee arthroplasty in non-cirrhotic patients7.
The goal of the present study was to evaluate the results of total hip and knee arthroplasty among patients with seropositive, but clinically asymptomatic, hepatitis C who were managed at our institution.
From 1995 to 2006, 7404 patients underwent primary total hip arthroplasty and 7391 patients underwent primary total knee arthroplasty at Thomas Jefferson University Hospital. Using our institutional database, we identified eighty-nine patients with hepatitis-C virus infection who underwent primary total hip or knee arthroplasty. The exclusion criteria were acute hepatitis, cirrhosis, known co-infection with hepatitis B or human immunodeficiency virus, coagulopathies such as hemophilia and von Willebrand disease, and revision joint arthroplasty. Patients were not routinely tested for human immunodeficiency virus or hepatitis B and were only excluded if their status with regard to these viral infections was known. Eighteen patients were excluded because of hemophilia, von Willebrand disease, or a history of co-infection with hepatitis B or human immunodeficiency virus. Seventy-one patients underwent seventy-two primary procedures (involving forty hips and thirty-two knees) and met the inclusion criteria. All of these patients had normal preoperative levels of liver enzymes. These patients were matched with a control group of patients who were seronegative for hepatitis-C virus. The patients in the control group were chosen randomly from our database and were matched with the patients in the study group for age, body-mass index, sex, year of surgery, and medical comorbidities (including diabetes, rheumatoid arthritis, and immunosuppressive conditions) in a 2:1 ratio.
The preoperative and postoperative clinical and radiographic records were reviewed for the demographic characteristics of the patients, details of the operative procedure, the method of contraction of hepatitis C, and the serological markers used to confirm the diagnosis. Intraoperative and postoperative complications, including medical complications (cardiovascular, neurologic, pulmonary, or gastrointestinal complications) and surgical complications (wound-healing problems, periprosthetic fracture, periprosthetic infection, instability, and loosening), reoperations, and revisions were studied in detail. The medical records of the patients in the control group were reviewed to ensure that these patients were not diagnosed with hepatitis-C virus.
The institutional review board approved this study.
Total Hip Arthroplasty Cohort: Preoperative and Operative Data (Table I)
The total hip arthroplasty cohort included twenty-nine male patients (one of whom had a staged bilateral total hip arthroplasty) and ten female patients with hepatitis C. The mean age at the time of surgery was fifty-five years (range, thirty-seven to eighty-nine years). The mean body-mass index was 29.1 kg/m2 (range, 18.9 to 40.8 kg/m2). The diagnosis of hepatitis C had been made at a mean of seventeen years (range, five to thirty-six years) prior to joint arthroplasty. In this cohort, thirty-eight total hip arthroplasties were performed for the treatment of osteoarthritis; one, for the treatment of sequelae of Legg-Calvé-Perthes disease; and one, for the treatment of a femoral neck fracture associated with degenerative joint disease. All patients underwent a primary total hip arthroplasty through a modified Hardinge approach and received standardized preoperative and postoperative care. One patient received general anesthesia, and the other patients received spinal anesthesia. All patients received uncemented acetabular and femoral implants.
The control group for the total hip arthroplasty cohort included sixty male and twenty female patients. The mean age at the time of surgery was 58.4 years (range, forty to eighty-eight years). The mean body-mass index was 28.3 kg/m2 (range, 17.8 to 38.2 kg/m2). All of the patients were diagnosed with degenerative joint disease. All patients had the same surgical approach, implants, and rehabilitation program as those in the study group.
Total Knee Arthroplasty Cohort: Preoperative and Operative Data (Table I)
The total knee arthroplasty cohort included fifteen male and seventeen female patients with hepatitis C. The mean age of the patients at the time of surgery was fifty-eight years (range, thirty-eight to seventy-four years). The mean body-mass index was 32.6 kg/m2 (range, 18.6 to 47.5 kg/m2). The diagnosis of hepatitis C had been made at a mean of thirteen years (range, five to twenty-four years) prior to joint arthroplasty. All of the patients were diagnosed with degenerative joint disease. All patients underwent primary total knee arthroplasty with the use of cemented posterior-stabilized components with a cemented patellar button. A medial parapatellar surgical approach was used in all knees. All patients received standardized preoperative and postoperative care.
The control group for the total knee arthroplasty cohort included thirty male and thirty-four female patients. The mean age at the time of surgery was sixty years (range, forty to seventy-six years). The mean body-mass index was 32.5 kg/m2 (range, 19.8 to 48.4 kg/m2). All of the patients were diagnosed with degenerative joint disease. All patients underwent a primary total knee arthroplasty with the use of cemented posterior-stabilized components with a cemented patellar button. A medial parapatellar approach was used in all knees. All patients received standardized preoperative and postoperative care.
Statistical Analysis
As the present study followed a group of patients and did not test a hypothesis, we did not consider statistical power for enrollment. The unpaired independent t test was used to compare the mean with variances among all study outcome parameters. The chi-square test and the Fisher exact test were used to compare the nonparametric outcomes. The level of significance was set at p < 0.05. All data were analyzed with use of SPSS software (version 13; IBM, Chicago, Illinois).
Source of Funding
We did not receive any outside funding in support of this study.
In the hepatitis-C group, the mean duration of follow-up was 101 months (range, sixty-six to 140 months) for the patients managed with total hip arthroplasty and 117 months (range, sixty-seven to 150 months) for those managed with total knee arthroplasty. In the control group, the mean duration of follow-up was ninety-four months (range, forty-five to 131 months) for the patients managed with total hip arthroplasty and ninety-eight months (range, forty-nine to 133 months) for those managed with total knee arthroplasty. There were other differences between the hepatitis-C group and the control group, including higher rates of autologous transfusion for the control group in both the total hip and total knee arthroplasty cohorts.
Total Hip Arthroplasty Cohort (Table II)
The perioperative hemoglobin drop, blood transfusion needs, and platelet counts for the study and control groups are presented in Table II. One patient sustained a femoral artery injury intraoperatively, which resulted in the need for revascularization, fasciotomy, and soft-tissue reconstruction surgery requiring skin grafting. This patient needed transfusion of two units of autologous blood and five units of homologous blood. This outlier was excluded in the statistical comparison of the two cohorts. In this cohort, one patient received ten units of platelets during admission with the diagnosis of thrombocytopenia.
Two patients (5%) developed supraventricular tachycardia. Three patients (7.6%) had persistent wound drainage requiring irrigation and debridement of the surgical wound, which resulted in a longer hospital stay (range, five to fourteen days). Three hips (7.5%) had surgical wound drainage, which was treated with oral antibiotics. The hospital course for the other patients was uneventful.
In the control group, two patients (2.5%) developed a urinary tract infection that was treated with oral antibiotics and one patient (1.25%) developed a foot drop that was treated with bracing. The latter patient recovered full function by six months postoperatively. Two patients (2.5%) had persistent wound drainage requiring treatment with oral antibiotics, and one patient (1.25%) had wound drainage requiring irrigation and debridement of the wound. The hospital course for the other patients was uneventful.
The prevalence of late complications was also higher in the hepatitis-C group. One hip (2.5%) had femoral implant loosening and one hip (2.5%) dislocated secondary to migration of the acetabular implant with loosening and was treated with revision total hip arthroplasty. One patient who had staged bilateral total hip arthroplasty had dislocation of one hip (2.5%) one month after surgery and dislocation of the contralateral hip (2.5%) forty-two months after surgery. The patient ultimately was managed with bilateral revision arthroplasty for the treatment of recurrent dislocations. The patient subsequently developed an infection in one hip that necessitated subsequent surgical procedures. In the control group, one patient (1.25%) sustained a periprosthetic femoral fracture and underwent revision total hip arthroplasty. One patient (1.25%) had a fractured femoral stem and was managed with revision total hip arthroplasty. One patient (1.25%) sustained multiple hip dislocations and was managed with revision surgery. One patient (1.25%) was readmitted to the hospital because of a cerebrovascular accident less than three weeks after surgery. The rate of revision in the control group was 3.8%, which was lower than the 10% rate of revision surgery for the hepatitis-C group (p = 0.167). The rate of dislocation in the control group was also lower than that in the hepatitis-C group (1.2% compared with 7.5%; p = 0.1).
To summarize the significant differences in complications, patients in the hepatitis-C group had a longer hospital stay (p = 0.001), a higher rate of persistent wound drainage (p = 0.03), and a higher rate of reoperation (p = 0.03).
Total Knee Arthroplasty Cohort (Table II)
The perioperative hemoglobin drop, blood transfusion needs, and platelet counts for the hepatitis-C and control groups are presented in Table II. One patient (3.1%) received fifteen units of platelets during the admission because of thrombocytopenia. One patient (3.1%) developed pulmonary emboli, which were treated with intravenous heparin and therapeutic Coumadin (warfarin). One patient (3.1%) had wound drainage and was managed with oral antibiotics. The hospital course for the other patients was uneventful.
In the control group, one patient (1.6%) developed an ileus and another patient (1.6%) developed hypotension; both patients were managed medically. One patient (1.6%) had an anaphylactic reaction to an undetermined substance but was managed successfully. Three patients (4.7%) had surgical wound drainage and were managed with oral antibiotics (p = 0.72). The hospital course for the other patients was uneventful.
In the hepatitis-C group, one patient (3.1%) required manipulation under anesthesia for the treatment of knee stiffness. Two patients (6.3%) underwent revision because of component loosening, and one patient (3.1%) fell and sustained a periprosthetic fracture that necessitated revision surgery. The latter patient subsequently developed a substantial extensor lag. In the control group, three patients (4.7%) had continued wound drainage postoperatively and were managed with oral antibiotics. One patient (1.6%) underwent manipulation under anesthesia for the treatment of knee stiffness. One patient (1.6%) developed a deep periprosthetic infection that was treated with a two-stage revision, and one patient (1.6%) underwent revision for the treatment of a mechanical problem.
To summarize the significant differences in complications, patients in the hepatitis-C group had a longer hospital stay (p = 0.01).
Combined Cohorts (Table III)
We combined the total hip and knee arthroplasty cohorts and compared all of the patients in the hepatitis-C group with those in the control group. To summarize the significant differences in terms of complications, the patients in the hepatitis-C group had a longer mean hospital stay and increased rates of all complications, surgical complications, mechanical complications, reoperations, and revisions in comparison with the control group. The odds ratios from univariate regression analysis are provided in Table IV.
Hepatitis C is a common disease, with a prevalence of 1.8% in the United States1. A substantial number of seropositive but non-cirrhotic patients undergo total joint arthroplasty. Little has been published on the outcome of total joint arthroplasty in patients with hepatitis C. The available reports in literature describe patients with hepatitis-C virus and co-infection with hepatitis-B virus or human immunodeficiency virus. The present study excluded patients with the human immunodeficiency virus or the hepatitis-B virus in an attempt to isolate the influence of seropositive, non-cirrhotic hepatitis C on the morbidity associated with total joint arthroplasty. We did not include patients with active hepatic disease. No patient with cirrhosis was included, and all patients with hepatitis C had normal levels of liver enzymes.
Although the number of patients was relatively small, the present study represents what we believe to be the largest cohort of non-cirrhotic patients with hepatitis-C virus who underwent total hip and knee arthroplasty that has been reported in the English-language literature. Patients with non-cirrhotic hepatitis-C virus infection who undergo total joint arthroplasty are at significantly increased risk of a longer hospital stay, complications, reoperation, and revision surgery. Patients with hepatitis-C virus who undergo total hip arthroplasty are at significantly increased risk of a longer hospital stay, persistent wound drainage, and reoperation. Patients with hepatitis-C virus who undergo total knee arthroplasty are at significantly increased risk of a longer hospital stay.
Hepatitis-C infection is frequently asymptomatic, and manifestations of the disease may be latent. However, hepatitis-C virus infection may be associated with multiple extra-hepatic diseases, such as thyroiditis, diabetes, thrombocytopenia, glomerulonephritis, inflammatory myositis, arthralgia, mixed connective tissue disease, leukocytoclastic vasculitis, and lymphadenopathy1,4,11. Most of these conditions are secondary to autoantibodies, cryoglobulins, and impaired lymphoproliferation, and their prevalence is increased among patients with long-standing infections1,4. Some of these immune-mediated manifestations, found in subclinical cases of hepatitis C, may influence the outcome of surgical interventions by altering the physiologic response to surgery. This phenomenon may result in higher rates of wound-healing problems and infection. In addition, hepatitis-C virus, even in the absence of cirrhosis, impairs platelet function and induces thrombocytopenia1,12,13. The role of platelets in hemostasis and the inflammatory response after surgery is essential, and any impairment in platelet function or quantity can result in increased bleeding and wound-healing problems and subsequently can increase the risk of infection. We found that patients with hepatitis C who underwent total hip arthroplasty had a significantly higher prevalence of wound drainage as compared with the control patients (p = 0.03). We believe that the higher prevalence of wound-related problems among the patients managed with total hip arthroplasty might be secondary to changes in platelet count and function. In contrast to the findings in the total hip arthroplasty cohort, the total knee arthroplasty cohort experienced no significant increase in the prevalence of wound-healing problems. This finding could be secondary to the use of a tourniquet, which may result in better small artery and capillary system hemostasis during the procedure, despite platelet dysfunction and thrombocytopenia.
One of the possible consequences of hepatitis-C infection may be an increased prevalence of diabetes mellitus. It has been hypothesized that patients with the hepatitis-C virus may have beta-islet cell dysfunction and subsequently may develop diabetes mellitus1,11. This, in turn, may result in an increased prevalence of wound complications and the potential for infection. In the present study, the prevalence of diabetes mellitus among patients with hepatitis C was 21% in the total hip arthroplasty group and 22% in the total knee arthroplasty group. These rates are considerably higher than the reported 10.1% prevalence of diabetes in the general population in the United States14. Because of matching the patients for comorbidities, including diabetes, the prevalence of this condition was similar in the hepatitis-C and control groups, and we are therefore unable to comment on the importance of this theory.
The present study did not find an increase in the rate of medical complications in the hepatitis-C group as compared with the control group. A possible explanation for this finding may be that, because of the disease, patients with hepatitis C were subjected to a more stringent medical workup and optimization prior to total joint arthroplasty. This may have had the effect of minimizing the prevalence of medical complications. Another possible explanation is that the small sample size of the cohort may have resulted in a type-II statistical error.
A limitation of this study is that it was retrospective. Patients in the control group were not tested preoperatively for occult cases of hepatitis C, and limited radiographic data were available to determine the effect of hepatitis C on the relative rates of mechanical complications. In order to minimize the influence of confounding variables, the present study was designed as a case-control study in which the cohorts were matched for potential confounders. This led to an identification of control patients who had a slightly shorter duration of follow-up. In addition, the rate of autologous blood donation was higher among the control patients because of blood bank restrictions discouraging autologous donation by patients with human immunodeficiency virus and hepatitis-C infections.
On the basis of the findings of the present study, it appears that patients with asymptomatic hepatitis C are at increased risk for the development of surgical complications following total joint arthroplasty. It is not known whether the presence of hepatitis C alone or other socioeconomic factors that may be shared among patients with hepatitis C contributed to the higher rate of complications seen in this group. It is possible that patients with hepatitis C were from lower socioeconomic groups and may have had other comorbidities that were not known at the time of matching the cohorts. We did not gather the socioeconomic data on the cohorts for this study, and patients were not randomized for these variables. Nonetheless, the study showed that, despite this possibility, patients with hepatitis C appear to have a more protracted postoperative course with a longer length of hospital stay and a higher prevalence of postoperative complications, most notably wound drainage. Considering the higher rate of lower platelet count and platelet dysfunction among patients with hepatitis C based on the literature published11-13, bleeding time may be useful as a preoperative admission test to evaluate the platelet function in this group of patients.