Papers

Follow up policy after treatment for Hodgkin's disease: too many clinic visits and routine tests? A review of hospital records

BMJ 1997; 314 doi: http://dx.doi.org/10.1136/bmj.314.7077.343a (Published 01 February 1997) Cite this as: BMJ 1997;314:343
  1. J A Radford, senior lecturer in medical oncologya,
  2. A Eardley, clinical audit and quality assurance facilitatorb,
  3. C Woodman, professor of cancer epidemiologya,
  4. D Crowther, professor of medical oncologya
  1. a Cancer Research Campaign Department of Medical Oncology, Christie Hospital NHS Trust, Manchester M20 4BX
  2. b Centre for Cancer Epidemiology, Christie Hospital NHS Trust, Manchester M20 4BX
  1. Correspondence to: Dr Radford
  • Accepted 29 November 1996

Abstract

Objective: To examine the effectiveness of routine clinic review in detecting relapse after treatment for Hodgkin's disease.

Design: Review of hospital records.

Setting: Regional centre for cancer treatment and research.

Subjects: 210 patients with Hodgkin's disease recruited to a chemotherapy trial protocol between 1984 and the end of 1990 who had achieved a complete or partial remission after treatment.

Main outcome measures: The number of clinic visits made by patients over the period of observation, the number of relapses occurring during that time, and the route by which relapse was detected.

Results: The 210 patients generated 2512 outpatient reviews, and 37 relapses were detected. Thirty relapses (81%) were diagnosed in patients who described symptoms, which in 15 cases had resulted in an earlier appointment being arranged. In only four cases (11%; 95% confidence interval 4% to 25%) was relapse detected as a result of routine physical examination or investigation of a patient who did not have symptoms.

Conclusions: Relapse of Hodgkin's disease after treatment is usually detected as a result of the investigation of symptoms rather than by routine screening of asymptomatic patients. It is therefore proposed that the frequency of routine follow up visits should be reduced and greater emphasis placed on patient education. This should underline the importance of symptoms and encourage patients to arrange an earlier appointment if these develop.

Key messages

  • Follow up after treatment for Hodgkin's disease has several functions but detection of relapse is probably the most important

  • In Hodgkin's disease the relapse rate is maximal 12-18 months after the start of treatment but declines rapidly thereafter

  • Relapse is usually identified as a result of the investigation of symptoms rather than by routine screening of asymptomatic patients

  • Routine clinic visits should be reduced in frequency and far greater emphasis placed on patient education; this should underline the importance of symptoms and encourage patients to arrange earlier appointments if these develop

Introduction

After treatment for Hodgkin's disease it is routine for patients to be reviewed in the clinic on a regular basis. A main purpose is to detect relapse at an early stage, when the tumour burden is light and salvage therapy is considered to have the best chance of effecting a cure. To our knowledge, however, the effectiveness of routine clinic review in detecting recurrent Hodgkin's disease has not been examined. We report such a study in a cohort of 210 patients treated at this hospital.

Patients and method

Follow up

Between 1984 and the end of 1990, 254 patients referred to this hospital with high risk stages I and II (presence of B symptoms or bulky mediastinal tumour, or both) and stages III and IV Hodgkin's disease were randomised in a collaborative phase III trial comparing mustine, vinblastine, procarbazine, and prednisone chemotherapy with a seven drug hybrid regimen (chlorambucil, vinblastine, procarbazine, and prednisolone and etoposide, vincristine, and doxorubicin). After chemotherapy patients in complete or partial remission received radiation treatment to sites of initial bulk disease or to areas of residual abnormality on the restaging computed tomography scan. Thirty patients either did not respond, had progression of their disease, or died during treatment and are not considered further. Full details of chemotherapy and radiotherapy, remission rates, causes of early death, progression free survival, and survival have been reported.1

The frequency of follow up visits for patients in remission after initial treatment accorded with current Manchester Lymphoma Group policy and Cotswold committee recommendations2–namely, two monthly visits in year 1, three monthly visits in year 2, four monthly visits in year 3, six monthly visits in years 4 and 5, and yearly visits thereafter. On each occasion patients are asked about symptoms and examined for evidence of recurrent disease. Blood is drawn for a blood count, the erythrocyte sedimentation rate, and a serum biochemical profile, and if the patient had mediastinal disease at presentation a chest radiograph is obtained. Other investigations are arranged as clinically indicated.

We examined the case notes of 210 of the 224 patients who achieved complete or partial remission with chemotherapy alone or chemotherapy plus radiotherapy and who were therefore eligible for outpatient review. Of the remaining 14 patients, eight lived far from the hospital and had been followed up by a local physician, one had emigrated, and case notes for the remaining five were unavailable.

The number of clinic visits during each of the first five years of follow up for 210 patients was recorded by a data abstractor and the case notes of patients who had relapsed put aside for further scrutiny. The second phase of the study was performed by three of us (JR, AE, and CW), who identified the following information for each relapsing patient: (a) the clinic visit that led directly to the diagnosis of relapse; (b) whether that visit was a routine appointment or one arranged by the patient, a relative, or another doctor because of the onset of symptoms; (c) whether it was a symptom volunteered by the patient, a sign identified on physical examination, or the result of a routine investigation in an asymptomatic patient that led directly to the diagnosis of relapse.

Risk of relapse

Estimates of how relapse rates varied over time were obtained by kernel smoothing3 of data for 369 patients achieving complete or partial remission after treatment in the collaborative trial1 from which the 224 patients for this study were derived.

Result

Outpatient review and relapse

Over the period of observation the 210 patients generated 2512 outpatient reviews (1086 in year 1, 667 in year 2, 420 in year 3, 231 in year 4, 108 in year 5), and 37 relapses were detected–a ratio of one relapse detected for every 68 visits. Thirty relapses (81%) were diagnosed in patients who described a symptom or symptoms which in 15 cases had resulted in an earlier appointment being arranged either by the patient or by a close relative (n = 10) or by a local doctor (n = 5). Thirteen of 30 patients reported one symptom, 10 patients reported two symptoms, and seven patients reported three or more symptoms (maximum five). Presence of a lump (n = 10), cough (n = 7), night sweats (n = 6), and weight loss (n = 6) were the most frequently reported symptoms (table 1).

Table 1

Symptoms reported by 30 patients found to have recurrent Hodgkin's disease

View this table:

In four cases (11%; 95% confidence interval 4% to 25%) relapse was detected as a result of routine physical examination or investigation of a patient who did not have symptoms (palpable lymphadenopathy (two cases), abnormal chest radiograph (two cases)). In a further three patients the case notes were unclear whether symptoms had been present. Figure 1 summarises the relation between relapse, presence or absence of symptoms, and whether or not the clinic visit was as previously arranged or early.

Fig 1
Fig 1

Flow chart showing relation of relapse, presence of symptoms, and visits to clinic for 210 patients followed up after treatment for Hodgkin's disease ‡Presence or absence of symptoms unclear from case notes in three cases.

Risk of relapse

Figure 2 shows the variation of relapse rates over time in 369 patients. Risk of relapse was not uniform over the period of follow up with a peak of 10 or 11 events per 100 patients yearly occurring 12-18 months after the start of treatment with a rapid decline to about five relapses per 100 patients yearly by year 3 and a further fall to fewer than two relapses per 100 patients yearly by year 5.

Fig 2
Fig 2

Kernel smoothed estimates of relapse rates over time for 369 patients in complete or partial remission after treatment for Hodgkin's disease. Epanechnikov kernel and a nine month bandwidth were used

Discussion

Routine review after treatment for Hodgkin's disease has several functions, including patient reassurance, detection of second tumours, and monitoring the late effects of chemotherapy and radiotherapy on gonadal, cardiac, and pulmonary function. Probably the most important function, however, is the early diagnosis of relapse. In this study 37 relapses were detected in 210 patients by using a follow up policy that generated a total of 2512 hospital visits during the period of observation. Of particular interest, 30 (81%) of the 37 patients in whom relapse was diagnosed had symptoms and in only four cases (11%) was relapse detected in an asymptomatic patient either by physical examination (two cases; 5.4%) or by a routine screening test (two cases; 5.4%). Furthermore, 15 of the 30 patients with symptoms either arranged an earlier clinic appointment for themselves or sought the advice of a local doctor, who arranged an earlier appointment on their behalf.

These findings are closely similar to those of Weeks et al, who studied follow up policy after chemotherapy for large cell lymphoma.4 They found that 32 of 36 relapses (89%) were detected because of symptoms and that only two relapses (5.6%) were diagnosed from an abnormal screening test result in an asymptomatic patient. A notable difference between the two studies was that in the American series all 32 patients with symptoms arranged an earlier clinic visit whereas in the British population only half did so. This may suggest either that American patients are better informed about the significance of symptoms and what action to take or that British patients are more reticent and less inclined to “trouble the doctor.” Nevertheless, it seems that a substantial proportion of patients already take action in response to symptoms during the follow up period.

The risk of relapse is maximal in the first two years after starting treatment (fig 2), which provides some basis for concentrating follow up effort during this period. Nevertheless, the absolute number of relapses is small (maximum 10 or 11 per 100 cases observed for one year), and in this study most were identified not as a result of routine screening of asymptomatic patients but by the appropriate investigation of symptoms when these occurred. Possibly a more intensive programme of surveillance (including interval computed tomography or gallium-67 scanning, for example) would lead to the earlier detection of relapse, but even if this was the case, benefit to patients would follow only if there was an associated survival advantage. Further studies are needed to address this issue and assess the degree of anxiety generated by or reassurance derived from a patient's visit to hospital for routine follow up.

In summary, our results suggest, firstly, that current follow up policy results in too many routine clinic visits; secondly, that relapse is usually diagnosed as a result of appropriate investigation of symptoms rather than by routine screening; and, thirdly, that at least some patients are prompted by symptoms to arrange an earlier appointment. A more rational strategy would be for routine visits to remain focused on the early years of follow up but to be reduced in overall frequency (we propose three monthly visits in year 1, four monthly visits in year 2, six monthly visits in year 3, and then yearly visits), coupled with far greater emphasis on patient education. This might include an information sheet to emphasise the importance of symptoms and encourage patients to arrange an earlier clinic appointment if they were, for example, to find a lump, lose weight, or develop a persistent cough or night sweats. We believe that such a policy would reduce the number of hospital visits but still provide sufficient opportunity for matters of concern to patients to be discussed and maximise the chances of early detection of relapse by encouraging patients to report new symptoms without delay.

Acknowledgments

The results of this study were presented at the 31st annual meeting of the American Society of Clinical Oncology, Los Angeles, 20-23 May 1995.

Funding: None.

Conflict of interest: None.

References

  1. 1.
  2. 2.
  3. 3.
  4. 4.
View Abstract