Abstract
Background
Measurement of plasma-free metanephrine plus normetanephrine (PFM) is the best screening test for phaeochromocytoma. While clearly raised levels are diagnostic, borderline increases may be associated with factors such as stress and medications, and should prompt a repeat study after interfering factors are withdrawn.
Methods
PFM results reported by a teaching hospital laboratory over a 12-month period were extracted from the laboratory information system. All borderline raised results were examined for a subsequent repeat test (as recommended by attached interpretative comment) and those not repeated were followed up by examination of case-notes or questionnaires to doctors.
Results
Of 111 patients with borderline increased PFM which did not normalize on repeat measurement, 33 were from the hospital and 78 from outside locations. Hospital notes for 27 out of 33 hospital-patients (82%) could be reviewed, and 49 completed questionnaires (63% of 78 sent out) were received from outside locations. Of these 76 patients thus followed up, the test was not repeated in 55 (72%) cases with borderline increased PFM. Of 10 patients with an adrenal mass and borderline PFM, only three had PFM repeated. Of another nine patients with undetermined final diagnosis and borderline PFM, only three had the test repeated.
Conclusion
Seventy-two per cent of borderline increased PFM results were not followed up with appropriate repeat testing, potentially leading to missed detection of phaeochromocytoma. A stronger interpretative comment may encourage appropriate repeat testing in more cases with borderline increased PFM and suspected phaeochromocytoma.
Introduction
The diagnosis of phaeochromocytoma depends importantly on biochemical evidence of increased production of catecholamines, usually achieved by analysis of plasma and urinary free catecholamines and catecholamine metabolites. Measurements of PFM provide the best test for excluding or confirming phaeochromocytoma due to high sensitivity and specificity. 1–6 A normal PFM result almost excludes phaeochromocytoma and no further testing is warranted. Moreover, in patients with large tumours, the PFM result is straightforward as it is invariably more than four-fold the upper reference limit, and thus the next direct step is to locate the tumour. However, patients with borderline elevated PFM results may present a diagnostic dilemma. This group of patients requires further investigations to either exclude or to confirm the presence of a tumour, as there are studies showing that 20–30% of patients with phaeochromocytoma have borderline-elevated PFM results. 2,3 It is important to consider the consequences of missing a diagnosis when taking into consideration the possibility of false-positive results in highly sensitive tests. Therefore, to improve the specificity of this test, it is recommended to do repeat testing after elimination of potential medications or factors that can cause false-positive results, rather than further investigation such as imaging studies. 1,5,7
This audit aimed to assess the management of patients with borderline-elevated PFMs results in order to find ways of improving the investigation of such patients in future in order to exclude or confirm phaeochromocytoma.
Methods
Patient results for PFM between 1 January 2008 and 31 December 2008 were extracted from the laboratory information system. Plasma concentrations of normetanephrine and metanephrine were determined in our laboratory using high-performance liquid chromatography based on the method of Lenders et al. 8
A borderline PFM result was defined as an increase of more than the upper reference limit but less than three times the upper reference limit. For patients rested 30 min with an indwelling cannula, this equates to normetanephrine between 610 and 2190 pmol/L and metanephrine between 310 and 1200 pmol/L. For seated patients without indwelling cannula or where collection conditions are not specified, this equates to normetanephrine 780–2190 pmol/L and metanephrine 300–1200 pmol/L. The laboratory results were accompanied with a general comment listing the possible factors that may cause the borderline-raised results (Table 1). The report also suggested that PFM testing should be repeated two weeks after withdrawal of the possible interfering factors.
Plasma-free metanephrine comment for borderline elevated results
The audit was performed in two parts: case-notes were reviewed for patients seen or warded at Royal Perth Hospital, and a questionnaire was sent to the managing doctor for patients managed in general practice or at other hospitals/clinics (see Appendix). We excluded patients who underwent repeat testing and patients from outside of Western Australia. The study protocol was approved by the Royal Perth Hospital Clinical Safety and Quality Unit, QI Registration No. 090603-2.
Results
Out of a total of 618 requests, there were 143 entries categorized as borderline PFM results of which nine were repeated subsequently with normal results. The remaining 134 entries included 23 repeat measurements; therefore, 111 patients were followed up further. Of these, 76 (68%) were able to be tracked: 27 out of 33 patients (82%) warded at the Royal Perth Hospital had their information obtained from case-notes, and 49 out of 78 doctors from outside locations (63%) responded to the questionnaire. The results were combined to give overall percentage responses. Key findings of the survey:
Forty-eight per cent of requests came from specialist clinics (28 government-based clinics and 20 private clinics), and 30% from general practitioners, and the rest (22%) were inpatients (Figure 1a). The main indication (54%) for requesting the test was for evaluation of hypertension (Figure 1b). Sixty-seven per cent of respondents to the questionnaire found the interpretative comment in the report either helpful or very helpful (Figure 1c). Seventy-two per cent of all patients able to be followed up did not have a repeat test (Figure 1d). (a) Locations of requesting doctors for plasma-free metanephrines (PFM) with borderline results are shown. Thirty percent and 28% requests came from general practitioners and specialist clinics, respectively. (b) Indications for requesting PFM among the borderline results are shown. The main indication was for evaluation of hypertension. (c) The perception of the doctors of the usefulness of the interpretative comment attached to PFM results for patient management. (d) The proportion of patients with borderline result who had PFM repeated (or not) as advised in the laboratory report
Of concern are two groups of patients in whom the test was not repeated: patients with adrenal mass and patients with undetermined diagnoses.
Borderline PFM results in patients with adrenal mass
Ten out of the 76 patients with borderline PFM results had an adrenal mass, and only three of them had the metanephrine test repeated. Of the seven cases not repeated, two had appropriate reasons: one patient had the blood sample taken postsurgery for elective cholecystectomy and adrenalectomy, a stressful condition that would have contributed to the raised PFM and, furthermore, the histology report for the adrenal tissue was negative for phaeochromocytoma. The second case was a patient with acromegaly, and the attending doctor had appropriately sent the patient for a clonidine suppression test. The other five patients with adrenal mass in whom the test was not repeated (Table 2) had imaging investigation done or planned instead of a repeat PFM test.
Patients with an adrenal mass and borderline increased plasma-free metanephrines that was not repeated
CT, computed tomography; US, ultrasound
Borderline PFM results in patients with undetermined diagnosis
Of the nine patients with borderline metanephrine results, who had not had their diagnosis determined, in three cases the doctors claimed the tests were in fact repeated subsequently. In three out of the remaining six cases the stated reasons for not repeating the tests were: cessation of symptoms, patient refusal of further treatment and patient failure to reattend clinic on appointment, respectively (Table 3). No reason was given for not repeating the test in the final three cases (Table 3).
Patients with an indeterminate final diagnosis and borderline increased plasma metanephrines that was not repeated
CT, computed tomography; US, ultrasound
Discussion
This limited retrospective audit confirms that the majority of doctors managing patients with borderline raised PFM do not repeat the tests as suggested. The majority of doctors surveyed found the comments helpful or very helpful, and yet they did not repeat the test as recommended.
PFM is a relatively new test offered by only a few centres in Australia. Many physicians are not yet familiar with this test; many still rely on urinary catecholamines as the initial screening test. Added to this is the low incidence of phaeochromocytoma, possibly contributing to most clinicians' lack of experience with the use of PFM in testing for phaeochromocytoma. Although the prevalence of this tumour is very low, it is a correctable and potentially life-threatening cause of hypertension and deserves an appropriate screening work-up. PFM is recommended as the best and most cost-effective initial test to use for screening for phaeochromocytoma. 1,5,9 There is also a recommendation to stop relying on urine catecholamines as the initial biochemical diagnosis of this rare tumour. 6 PFM results within the reference interval indicate that phaeochromocytoma is highly unlikely and no further testing is needed, whereas a patient with suspected phaeochromocytoma and a borderline increased PFM result needs further evaluation. At that point, it is important to consider any potential associated factors or the influence of medications that may cause a false-positive test result, eliminate it, and repeat the test or consider additional biochemical tests most appropriate to firmly establish or refute the diagnoses of phaeochromocytoma. A normal PFM in a repeat measurement virtually rules out the diagnosis. Eisenhofer and colleagues 7,9 recommended the coupling of the clonidine suppression test with measurements of PFM in patients with persistently increased PFM in the borderline range. This strategy would help to minimize delay in diagnosing phaeochromocytoma and avoid costly and time-consuming imaging studies in patients who do not have the tumour.
In our audit, the diagnosis of phaeochromocytoma could have been missed in patients with borderline PFM results, especially in patients with adrenal masses or undetermined diagnosis, although even in the case of those given the label of essential hypertension, the single borderline increased PFM result may have contributed to excluding the diagnosis of phaeochromocytoma, especially if the pretest suspicion (probability) was low. In a study of 42 patients with incidentaloma and borderline elevated PFM, 14 (33%) were subsequently shown to have phaeochromocytoma. 10 The authors of that study concluded that clinical factors cannot distinguish between those with and without phaeochromocytoma in that group and advocated either routine alpha-blockade preoperatively or further diagnostic tests to better characterize the tumour. In our audit, in the five patients with adrenal mass in whom the test was not repeated, it would have been more cost-effective to have the PFM test repeated rather than having expensive imaging studies done.
A limitation of this audit is that it was laboratory-based and hence no detailed follow-up of patients was readily available. Another weakness of this study was the short duration of audit; some patients were still undergoing follow-up and repeat testing at the conclusion of our audit. It is also recognized that analysis of PFM will not necessarily identify phaeochromocytomas which predominantly secrete dopamine although these may well have altered morbidity and mortality compared with noradrenaline and adrenaline secreting phaeochromocytomas.
We hypothesize that the interpretative comment accompanying borderline PFM results may have misled doctors into thinking that the report ruled out phaeochromocytoma, even though in fact the report indicated that the presence of tumour was still possible and these patients needed further tests to refute or confirm its presence (Table 1).
This audit suggests that the interpretative comment for borderline PFM results should be rephrased to stress the fact that phaeochromocytoma may present with borderline elevated PFM and to more strongly recommend repeat testing (Table 4). Physician education on the use of PFM as screening for phaeochromocytoma may also be of benefit, particularly to the non-specialist. It is hoped that the correct use of PFM as the initial screening test and appropriate follow-up will simplify and improve the diagnosis of phaeochromocytoma.
A suggested alternative comment for borderline elevated plasma free metanephrine results
Footnotes
DECLARATIONS
Appendix
GP, general practitioner
AUDIT OF MANAGEMENT OF BORDERLINE RAISED PLASMA-FREE METANEPHRINES
QI Registration No.: 090603-2
1
Patient's details
Surname:
First name:
Medical Record Number (MRN):
Date of birth:
2
GP
□
Endocrinologist
□
Surgeon
□
Registrar… … … … … … …(please specify)
□
Others… … … … … … …(please specify)
□
3
Clinic:
Specialist
□
Private practice
□
GP
□
Ward:
□
4
Evaluation of hypertension
□
Unexplained spells (headache, palpitations, light headedness, excessive perspiration, pallor, flushing)
□
Incidental adrenal masses
□
Family history of pheochromocytoma
□
Others (please specify)… … … … … … … … … … … … …
□
5
… … … … … … … … … … … … … … … … … .
6
Medical
□
Surgical
□
Combination
□
7
Very helpful Helpful Slightly helpful Unhelpful
Neither helpful nor unhelpful
□ □ □ □
□
8
Yes No
If no, please state reason why
□ □
… … … … … … … … … …
9
Imaging
□ (please specify)… … … … … … … …
Others
□ (please specify)… … … … … … … …
10
In relation to question 9, what were the findings: