 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1998 The Journal of Bone and Joint Surgery, Inc. Total Knee Arthroplasty in Morbidly Obese Patients* , PHILADELPHIA, PENNSYLVANIAInvestigation performed at the Department of Orthopedics, University of Pennsylvania, Philadelphia
We reviewed the clinical outcomes of fifty primary total knee arthroplasties that had been performed with cement in forty patients who were considered morbidly obese (a Quetelet index of more than forty). These results were compared with those of 1768 similar procedures, performed during the same time-period by the same surgeon, in 1539 patients who were not morbidly obese (controls). At a mean of approximately five years postoperatively, there was a significant difference between the morbidly obese patients and the control group with regard to the knee and functional scores (84 and 53 points compared with 92 and 67 points; p < 0.00005 for both scores). No significant difference was detected, with the numbers available, with regard to the range of motion or the radiographic score (p = 0.77). The rate of perioperative complications was significantly higher in the morbidly obese patients (p < 0.00005). Of the fifty knees in these patients, eleven (22 percent) had a wound complication, five (10 percent) had an infection, and four (8 percent) had an avulsion of the medial collateral ligament. The five infections developed within twenty weeks after the operation, and three were associated with a wound complication. In comparison, thirty-five (2 percent) of the 1768 knees in the control group had a wound complication, eleven (0.6 percent) had an infection, and none had an avulsion of the medial collateral ligament. We concluded that total knee arthroplasty in morbidly obese patients can be successful but is associated with an increased rate of perioperative complications, including problems with wound-healing, infection, and avulsion of the medial collateral ligament. Alterations in the operative technique for soft-tissue closure and protection of the medial collateral ligament have decreased the rates of complications related to wound-healing and the medial collateral ligament.
Obesity has been associated with both degenerative joint disease6,11,14,17,20,32,38 and an increased predisposition to meniscal injuries of the knee13. Thus, many patients who have a total knee arthroplasty are obese6,13,14,20,34, and a subset is considered morbidly obese. Although there have been several reports in the literature on the outcome of total knee arthroplasty in obese patients1,2,10,24,34,36, we are aware of none that have focused on morbidly obese patients. There are several definitions of morbid obesity7. In clinical studies, two methods are commonly used: estimation of relative weight (body weight divided by the midpoint of the desirable weight for a person of medium frame as recommended in the 1959 or 1983 Metropolitan Life Insurance Company tables23) and calculation of body-mass index, also called the Quetelet index7. The Quetelet index relates well with the body-fat percentage and is the preferred method for assessment of the magnitude of potential health risks associated with excessive body weight3,5-7,26. This index is derived by dividing an individual's weight (in kilograms) by his or her height (in square meters)7. Patients who have a body-mass index of twenty to thirty are considered of normal weight; thirty to forty, obese; and more than forty, morbidly obese7. For example, a patient who is 1.67 meters (five feet and six inches) tall must weigh at least 113 kilograms (250 pounds) in order to be considered morbidly obese. Morbidly obese patients are considered to be at high risk when they are managed with general surgical and gynecological operations26-29,35. There is a general consensus that excessive weight is a risk factor in total knee arthroplasty1,2,10,33,35. Stern and Insall studied the effect of weight on the outcomes at a mean of four years after fifty-four total knee arthroplasties in obese patients34. Those authors reported patellofemoral symptoms in sixteen knees (30 percent), but there were no other specific problems. Their study population included twelve patients who were severely obese (more than 175 percent of their ideal weight according to the 1983 Metropolitan Life Insurance Company table23). This subgroup probably weighed considerably less than the morbidly obese patients in our study. In order to better understand the potential problems associated with total knee arthroplasty in morbidly obese patients, we compared the outcomes of this procedure in patients who were and were not morbidly obese.
Morbidly obese patients (those having a Quetelet index of more than forty) were identified from a database containing information on all patients who had had a total knee arthroplasty performed by the senior one of us (P. L.) between January 1980 and September 1994 and had been followed for a minimum of two years. The study group consisted of forty patients (fifty knees). Five morbidly obese patients had had a bilateral total knee arthroplasty, with the procedure staged at intervals ranging from three months to three years. The control group consisted of the remaining 1539 patients (1768 knees) who had had the same procedure, performed by the same surgeon in the same time-period, with the same minimum duration of follow-up. The patients were evaluated preoperatively and postoperatively with use of the rating system of the Knee Society16. This system assigns a knee score that is based on an evaluation of the knee joint itself, including its range of motion, and a functional score that rates the patient's ability to walk and to climb stairs. Both scores have a range of 0 to 100 points. The scores that were recorded at each patient's most recent outpatient visit were used in this study. The various prostheses that were used during the fifteen-year interval of the study included the total condylar (Howmedica, Rutherford, New Jersey), posterior stabilized (Howmedica), Miller-Galante I (Zimmer, Warsaw, Indiana), press-fit condylar (Johnson and Johnson Orthopaedics, Braintree, Massachusetts), and press-fit condylar substituting (Johnson and Johnson Orthopaedics) designs. Each type of implant was used in at least 200 knees, and no single type was used preferentially in morbidly obese patients. Perioperatively, all patients were given antibiotics prophylactically. Aspirin was the medication used most commonly (1618 [89 percent] of the 1818 procedures) for prophylaxis against thromboembolism; it was given unless the patient was enrolled in an alternative research protocol with warfarin or low-molecular-weight heparin. Obese patients were not managed differently with regard to prophylaxis against thromboembolism. All fifty procedures were performed with use of a medial parapatellar exposure. An anterior medial incision was made through the skin, fat, and fascia to the anterior border of the quadriceps tendon and along the medial border of the patella and the patellar ligament22. Closure of the tendon and the quadriceps expansion was followed by closure of the fascia, careful subcutaneous closure, and closure of the skin with use of nylon interrupted sutures or staples. A suction drain was used until 1990 and then this practice was discontinued. Knee drains, when used, were removed on the first postoperative day. All intraoperative and postoperative complications were recorded in a database. Radiographs were evaluated at each postoperative visit, and those made most recently were analyzed for the current study. All radiolucent lines were recorded and were classified according to the scoring system of the Knee Society12. Wound complications were recorded carefully and were distinguished from deep infections of the knee. A wound complication was defined as separation of the skin margins with prolonged drainage necessitating an alteration in the normal postoperative course (immobilization of the knee or operative exploration of the wound). A Fisher exact two-tailed test was performed to analyze the significance of any difference, between the morbidly obese patients and the control group, with regard to age, Quetelet index, wound complications, rate of infection, and knee, functional and radiographic scores. P values of less than 0.05 were considered significant.
The mean age of the forty morbidly obese patients at the time of the operation was 64.6 years (range, 45.6 to 76.5 years). There were thirty-six women and four men. The preoperative diagnosis was osteoarthrosis in thirty-eight patients (95 percent), rheumatoid arthritis in two (5 percent), and Paget disease in one (3 percent). (One patient had both osteoarthrosis and Paget disease.) The mean Quetelet index was 44.0 (range, 40.3 to 56.2), with the men averaging 1.7 meters (range, 1.5 to 1.9 meters) in height and 113.3 kilograms (range, 97.2 to 158.9 kilograms) in weight and the women averaging 1.6 meters (range, 1.3 to 1.7 meters) in height and 112.3 kilograms (range, 78.5 to 136.1 kilograms) in weight. These patients were followed for a mean of 4.8 years (range, two to thirteen years). The mean age of the 1539 patients in the control group at the time of the operation was 67.7 years (range, 13.4 to 94.2 years). There were 1016 female patients and 523 male patients. The preoperative diagnosis was osteoarthrosis for 1432 knees (81 percent) and rheumatoid arthritis for 195 (11 percent); the remaining 141 knees (8 percent) had other diagnoses. The mean Quetelet index was 28.0 (range, 10.9 to 39.7), with the male patients averaging 1.8 meters (range, 1.5 to 2.0 meters) in height and 88.0 kilograms (range, 54.4 to 136.0 kilograms) in weight and the female patients averaging 1.6 meters (range, 1.4 to 1.9 meters) in height and 75.0 kilograms (range, 21.3 to 136.0 kilograms) in weight. Preoperatively, the morbidly obese patients had a mean knee score of 31 points (range, 0 to 66 points) and a mean functional score of 40 points (range, 0 to 70 points). Postoperatively, the mean knee score increased to 84 points (range, 0 to 100 points) and the mean functional score, to 53 points (range, 0 to 90 points). Postoperatively, the mean range of motion was 106 degrees (range, 45 to 125 degrees) and the mean radiographic score was 98 points (range, 70 to 100 points). Preoperatively, the patients in the control group had a mean knee score of 65 points (range, 0 to 100 points) and a mean functional score of 45 points (range, 0 to 100 points). Postoperatively, the mean knee score increased to 92 points (range, 38 to 100 points) and the mean functional score, to 67 points (range, 20 to 100 points). Both of these postoperative scores were significantly higher than those in the morbidly obese group (p < 0.00005). Postoperatively, the mean range of motion was 109 degrees (range, 15 to 130 degrees) and the mean radiographic score was 98 points (range, 60 to 100 points). With the numbers available, no significant difference between the groups could be detected with regard to either the range of motion or the radiographic score (p = 0.77). Problems with primary wound-healing (hereafter referred to as wound complications) and other perioperative problems were significantly more common in the morbidly obese patients. Of the fifty knees in those patients, thirteen (26 percent) had a perioperative complication compared with thirty-two (2 percent) of the 1768 knees in the control group (p < 0.00005). Eleven (22 percent) of the fifty knees had a wound complication and five (10 percent) had a deep joint infection, whereas in the control group a total of thirty-five (2 percent) of the 1768 knees had a wound complication and a total of eleven (less than 1 percent) had a deep joint infection. All five infections in the morbidly obese patients developed within twenty weeks after the operation, and three were associated with a wound complication. The differences in the rates of infection and wound complications between the two groups were highly significant (p < 0.0001 and p < 0.00005, respectively). There were no perioperative deaths and no fatal pulmonary emboli within three months after the operation in either group. All eleven wound complications in the morbidly obese patients were treated according to their extent and type. Only one such problem occurred in a patient who had type-II (non-insulin-dependent) diabetes. Wound drainage was treated with immobilization of the knee and oral administration of antibiotics. If drainage continued for more than six days or if there was any evidence of cellulitis, the wound was explored, irrigated, and reclosed. Separation of the wound was treated with packing and healing (by so-called secondary intention), or, if the wound was clean, it was closed by so-called tertiary intention. Two of the five infections in the morbidly obese patients developed within eight weeks after the operation and were treated with irrigation and débridement, with retention of the prosthesis. The remaining three infections were treated with removal of the implant and delayed exchange. Staphylococcus aureus was the cause of four of the five infections, and Staphylococcus epidermidis was the cause of the fifth infection. There was no evidence of recurrence of any of these five infections during the period of follow-up, and all five were considered to have been treated successfully. Two prostheses in morbidly obese patients failed because of mechanical reasons not associated with infection. The first failure occurred when a patient who had rheumatoid arthritis fell during the perioperative period, landing directly on the knee and sustaining a fracture of the lateral aspect of the tibial plateau. The second failure, which was due to aseptic loosening, occurred two years after the operation in a patient who had osteoarthrosis. Both of these patients had a revision. There were no intraoperative avulsions of the medial collateral ligament in the control group, whereas four morbidly obese patients had this complication. The avulsion, from the tibial insertion, was recognized at the time of the operation, when a difficult exposure became easy. Exploration of the ligament revealed the avulsion in all four knees. In three of them, a posterior-cruciate-substituting nonconstrained total knee prosthesis was used because the distal end of the ligament appeared unshredded and healthy and could be stapled back into place. In the fourth knee, a total condylar-III constrained implant (Johnson and Johnson Orthopaedics) was inserted. None of these patients needed a revision or had long-term instability, and all were doing well at the time of the latest follow-up.
Total knee arthroplasty is one of the most predictable and successful reconstructive procedures performed on adults8,30,31. A meta-analysis of 10,000 patients revealed an 89 percent rate of good and excellent outcomes at a mean of four years postoperatively8 and a 95 percent rate of survival of the prosthesis at ten to fifteen years postoperatively8,30,31. Although a high percentage of patients who have a total knee arthroplasty are obese6,13,20,34, we are not aware of any reports on the results of this procedure in morbidly obese patients. Therefore, the focus of the current study was the outcomes and complications of total knee arthroplasty in this unique population. Two major concerns with regard to morbidly obese patients who have a total knee arthroplasty are mechanical loosening and perioperative morbidity. Several authors have reported that increased body weight increases stresses at the cement-bone interface, potentially leading to a higher rate of aseptic loosening10,20,25. However, Stern and Insall, in a study of 182 total knee arthroplasties, including twelve (7 percent) performed in severely obese patients (those having more than 175 percent of their desired body weight, as defined by the 1983 Metropolitan Life Insurance Company table23), found no clear-cut increase in the rate of loosening34. This finding, which may have been related to the overall lower level of activity of these patients, is in contrast to the increased rates of loosening that have been noted after total hip arthroplasties in obese patients15,21. There may be a strong association between obesity and wound complications. Wong et al. prospectively studied the results of 121 total knee arthroplasties and found a linear relationship between obesity and the prevalence of wound complications40. Those authors found no relationship to other risk factors, such as rheumatoid arthritis, diabetes, smoking, age, hypertension, increased blood loss, and gender. Obese patients who had general surgical procedures, including gastric bypass operations, hysterectomies, cholecystectomies, and operations for the treatment of ulcers, have been noted to have increased rates of wound complications compared with patients who were of average weight26-29,35,37. Two generally accepted theories for this finding include a weakened immune response and poorly vascularized fatty tissues in obese patients9,26. In support of the weakened-immune-response theory, Krishnan et al. found that the number of monocytes that differentiated into macrophages in vitro was lower in obese subjects than it was in normal subjects19. Kolterman et al. found that the generation of lymphocytic migration-inhibiting factor in the presence of normal levels of glucose was significantly less in obese subjects than it was in controls (p < 0.001)18. Obesity also has been frequently associated with insulin resistance and hyperglycemia6,18, conditions known to interfere with leukocyte function. Complications related to wound-healing clearly contribute to the risk of infection and to major morbidity after total knee arthroplasty. Wilson et al. found that thirteen (19 percent) of sixty-seven infections at the sites of total knee arthroplasties in obese patients were associated with a wound complication compared with two (3 percent) of sixty-seven such infections in a control group39. In the current study, three of the five infections in morbidly obese patients were associated with a wound complication. Problems with wound-healing can be minimized. The major risk of these problems in morbidly obese patients was noted several years ago, and changes in operative technique were made in order to avoid them. During the last four years, we have not encountered any major wound complications or infections following total knee arthroplasty in this population. We believe that this is due to increased awareness of the risks as well as to changes in the management of the operative wounds. For morbidly obese patients, we use the following techniques. A previous scar, when present, is routinely excised down to the fascia. The incision is slightly longer than the one used in a routine total knee arthroplasty. Rubber-shod vertical mattress retention sutures are used in the areas with the most fat at the time of closure, with care being taken not to overtighten the sutures and to approximate the wound carefully with a multilayer closure. If any wound separation is noted, it is addressed immediately with exploration, irrigation, and reclosure in the operating room. The sutures are removed on the fifth postoperative day. In addition to complications related to wound-healing, there is the potential for avulsion of the medial collateral ligament during total knee arthroplasty in obese patients. Hyperflexion of the knee against an obese thigh, during an attempt to obtain better exposure, pushes the tibia against the rolls of fat, forcibly subluxating the tibia anteriorly. This increases the tension in the ligament and the risk of avulsion. Awareness of this potential complication, early removal of osteophytes (which cause tension in the medial collateral ligament), and avoidance of hyperflexion can help to prevent this problem. Proper alignment of the limb can also be difficult in morbidly obese patients because the fat makes effective exposure difficult. We recommend inverting the patella within the prepatellar bursa, beneath the fat. This allows better visualization of the corner of the lateral aspect of the tibia. Removal of approximately 50 to 70 percent of the retropatellar fat pad also increases the exposure of the lateral aspect of the tibia. As the fat pushes the extramedullary tibial guide medially, it may lead to varus malalignment. We recommend the use of intramedullary guides for both the tibia and the femur. If these are not available, the surgeon should be aware of this potential to osteotomize the tibia in varus. In addition, instruments used for exposure, such as z-retractors, may be too short and become lost in the fat. Bent Homan retractors, which are longer and have minimally curved tips, provide better retraction. Attention to details such as these will help to expedite the operation and to make the technical results more consistent. Although our study revealed an increased rate of perioperative complications in morbidly obese patients, it confirmed the results of other studies24,34 in which functional, range-of-motion, and knee-rating indices were found to be comparable with those for patients who were not obese. However, we found a significant difference in the functional and knee scores between the morbidly obese patients and the control group (p < 0.00005) in the current study; this difference was due to the lower level of activity of the former group. It has previously been shown that morbidly obese individuals have a considerably lower level of activity and energy expenditure4. In summary, the effect of morbid obesity on the outcome of total knee arthroplasty is of great concern to the orthopaedic surgeon. Proper operative technique, experience, and meticulous attention to wound-healing are important and are related to the overall success of the procedure in this population. We recommend that morbidly obese patients be made aware of their greater risk of complications following total knee arthroplasty and that the surgeon attempt to decrease this risk by exercising increased vigilance in the care of the wound during the perioperative period.
NOTE: The authors thank Dr. Allyson T. Gage for her contributions to the manuscript.
*One or more of the authors has received or will receive benefits for personal or professional use from a commercial party related directly or indirectly to the subject of this article. In addition, benefits have been or will be directed to a research fund or foundation, educational institution, or other nonprofit organization with which one or more of the authors is associated. Funds were received in total or partial support of the research or clinical study presented in this article. The funding source was Johnson and Johnson Orthopaedics.
  CiteULike  Connotea  Del.icio.us  Facebook  Technorati  Twitter What's this?
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
| home | CME | help | feedback | subscriptions | archive | site index |
|
Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online. |