Commentary & Perspective

Commentary & Perspective on
"The Fate of Patients Not Returning for Follow-up Five Years After Total Knee Arthroplasty"
by Paul J. King, MD, et al.

Commentary & Perspective by
Robert T. Trousdale, MD*,
Department of Orthopedics, Mayo Clinic, Rochester, Minnesota

Long-term clinical studies have become the "gold standard" with which to document the effectiveness of total knee arthroplasty. Inevitably, some patients are going to be lost to follow-up. The fate of these patients is important if an accurate assessment of the true outcome of knee arthroplasty is to be made. Some investigators have addressed the issue of statistical handling of patients lost to follow-up in clinical studies^1-4. Some have suggested that patients who were lost to follow-up do not have a statistically different outcome than those who were followed⁴. Others have refuted this concept and have suggested that the patients who were lost to follow-up can be expected to have a poorer outcome than those who were followed³.

In this study, King et al. have highlighted some very important points with regard to patient follow-up. They point out that the Internet can be a very effective, inexpensive, and efficient way to find patients whose most recent contact information at a doctor's office is inaccurate. These authors were easily able to identify 95% of the patients who were lost to follow-up. They provide helpful information by listing the Internet search engines that were used (ancestry.com, anywho.com, superpages.com, switchboard.com, people.yahoo.com, whitepages.com, and whowhere.com).

The authors also accurately document the reasons for loss of follow-up. They found that patients who were older at the time of knee surgery were more likely to be lost to follow-up. They also document the reasons that the patients gave for not returning for follow-up appointments. These included death of the patient, not having any problems with the knees, forgetting or not knowing that follow-up was recommended, feeling that the trip to the doctor's office was onerous, being too busy or preoccupied with other health issues, seeing another surgeon because of convenience, wanting to delay the perceived inevitable replacement of the other knee, and being unhappy with the care provided. This information is helpful as it may guide surgeons to educate patients, preoperatively and early in the postoperative period, about the importance of continuing follow-up.

Perhaps most importantly, this manuscript documents that patients who are lost to follow-up may not necessarily have had a poor outcome. The authors found that the patients who had been lost to follow-up had similar Knee Society pain and function scores and similar rates of reoperation when compared with the patients who had complied with a follow-up protocol.

Although the authors' conclusion that "patients who did not attend follow-up appointments in this series did not have any additional surgery and did not have any significant differences in measured outcome variables when compared with patients who had complied with a follow-up protocol" is well supported by the data, I do not think that these findings can be assumed to be universal. These data probably do not apply to all surgeons or to all types of knee implants. This paper reports on a group of patients who were operated on by an experienced, relatively high-volume knee surgeon and received a single implant design (PFC Sigma; DePuy, Warsaw, Indiana) that has had documented, clinical success. The fact remains that other researchers who have lost patients to follow-up cannot be 100% certain of the clinical and radiographic results of these patients unless they have "tracked" the patients down. Furthermore, surgeons who are not at a tertiary care center may have patients who are more likely to seek care elsewhere if they are not doing well. As suggested by Murray et al., it may be best for clinical researchers to include the lost-to-follow-up group in their results, first giving the data as if the lost patients did well (the best-case scenario) and then as if the lost patients did poorly (the worst case scenario)^2,3. Three data points would be provided, corresponding to the known follow-up group, the best-case scenario, and the worst-case scenario. Unless the outcome of all of the patients in the clinical follow-up series is truly documented (i.e., the surgeon and center have documented the outcome of the lost-to-follow-up group as these authors have), it remains impossible to know whether the outcome of patients who have been lost to follow-up can be associated with success or failure.

*The author did not receive grants or outside funding in support of his research or preparation of this manuscript. He did not receive payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity. No commercial entity paid or directed, or agreed to pay or direct, any benefits to any research fund, foundation, educational institution, or other charitable or nonprofit organization with which the author is affiliated or associated.

References

1. Twisk J, de Vente W. Attrition in longitudinal studies: how to deal with the missing data. J Clin Epidemiol. 2002;55:329-37.
2. Murray DW, Carr AJ, Bulstrode C. Survival analysis of joint replacements. J Bone Joint Surg Br. 1993;75:697-704.
3. Murray DW, Britton AR, Bulstrode CJ. Loss to follow-up matters. J Bone Joint Surg Br. 1997;79:254-7.
4. Joshi AB, Gill GS, Smith PL. Outcome in patients lost to follow-up. J Arthroplasty. 2003;18:149-53.

Copyright © 2004 by the The Journal of Bone and Joint Surgery, Inc.