Diagnostic tests in urology: percutaneous biopsy of renal masses

Authors

Errata

This article is corrected by:

  1. Errata: Corrigendum Volume 111, Issue 5, 852, Article first published online: 12 April 2013

Correspondence: Rodney H. Breau, Department of Surgery, Division of Urology, 501 Smyth Road, Box 222, Ottawa, ON, Canada K1H 8L6.

e-mail: rodneybreau@yahoo.ca

Abstract

What's known on the subject? and What does the study add?

  • Small renal masses are often detected incidentally with abdominal CT. About 80% of these masses are malignant. Many studies have investigated the diagnostic accuracy of small renal mass biopsy using various techniques. This has resulted in a broad range of published sensitivities and specificities without a clear guide for clinical use.
  • This evidence-based medicine article addresses the clinical utility of renal mass biopsy. Based on a common patient scenario, a guide to calculating the probability of malignancy after a benign biopsy is provided. The scenario is used to show the potential influence of renal mass biopsy on clinical decision-making. This appraisal also discusses the limitations of the current literature, highlighting the need for multi-institutional studies with standardised biopsy techniques and a reliable comparator to confirm the clinical utility of this diagnostic test.
Abbreviation
LR(−)

likelihood ratio (negative)

Case Scenario

You are reviewing new patient referrals before a clinic. Among them is a 55-year-old healthy man with a 3-cm solid enhancing renal mass (stage cT1a) on CT who has been referred to you for a second opinion. According to the note from your colleague, the patient is considering active surveillance. Included in the referral is the finding from an ultrasound-guided core biopsy of the mass that suggests oncocytoma. You have limited experience with interpretation of renal mass biopsies, as this is not a diagnostic test you have traditionally ordered in patients with radiographic images consistent with RCC. To accurately counsel your patient, especially considering the biopsy results, you decide to review the literature on percutaneous biopsy of small renal masses.

Clinical Question

How do the results of a percutaneous core biopsy of a small renal mass affect the probability that a patient will have pathologically confirmed malignancy after surgical resection?

Finding the Best Evidence

Using techniques outlined in a previous publication, you perform a literature search to find evidence to answer your clinical question [1]. You start by identifying the baseline prevalence (pre-test probability) of malignancy in patients with small renal masses. Then, using a systematic approach you search the literature to find the best evidence on the diagnostic performance of small renal mass biopsy [2].

The Evidence

Pre-Test Probability

You find that 20% of small, solid, enhancing renal masses on CT are benign tumours while the remaining 80% are malignant [3-7]. You also learn that younger patients, men and larger tumour size have a higher probability of malignant histology [5-9]. Using a nomogram that incorporates gender, age, and, tumour size, the calculated probability of malignancy is 86% in the referred patient before undergoing renal mass biopsy [5].

Sensitivity and Specificity

To evaluate the impact of ultrasound-guided biopsy on the probability of malignancy, you need to determine the likelihood ratios (LRs) associated with this diagnostic test. If not referenced, LRs may be calculated using the sensitivity and specificity of the test. You use the Pubmed Clinical Queries tool to search ‘renal cancer AND percutaneous biopsy’. Within the systematic reviews you find several articles on the management of small renal masses, but none provide enough information to calculate the sensitivity and specificity of biopsy. You then return to Clinical Queries and choose the diagnosis category. This search yields 386 articles that are reduced to 20 when you limit to ‘review articles; written in English; within the last 5 years’. One large review provides information on diagnostic performance, so you download it for appraisal [10].

The review summarises 27 articles that includes 2474 renal mass biopsies. The authors of the review article highlight the difference between a non-diagnostic biopsy (i.e. not enough tissue for the pathologist to evaluate) and an inaccurate biopsy, where enough tissue was obtained on biopsy, but the histological interpretation was different compared with the surgical tumour specimen. You are struck by the considerable variation between studies, as sensitivities ranged from 70% to 100% and specificities ranged from 60% to 100%. Potential explanations for the variability were differences between institutional techniques, biopsy needle sizes, and image guidance. You recognise the available data may also be biased, as many patients with a benign biopsy did not undergo tumour removal, thus it is difficult to determine the true incidence of ‘false negative’ biopsies. Acknowledging these limitations, you think the heterogeneous study characteristics may be a reflection of ‘real-world’ practice and the pooled performance from this review is the best estimate available to apply to your patient (sensitivity 92.1%; specificity 89.7%).

LRs

LRs allow you to calculate the increase or decrease in odds of a patient having a disease after performing a diagnostic test. In this case, the negative LR (LR–) allows you to determine how the negative (benign) biopsy decreases the probability that your patient's tumour is malignant. You calculate the LRs using the equation in the Appendix and determine that a negative biopsy has a LR– of 0.09.

Case Resolution – Applying the Results to the Care of Your Patient

You have gathered the necessary information to assess the probability that the 55-year-old gentleman's renal mass is malignant. Using an estimated pre-test probability of 86% and LR– of 0.09, you use an online calculator to determine the probability of malignancy [11]. The online calculator is simply a quick method to obtain post-test probabilities. The same results could be obtained from a Fagan nomogram (Fig. 1). The probability of malignancy in this man with a negative biopsy is 36%. The patient is appreciative of your risk assessment, as he was surprised that his chance of malignancy was as high as 36%. Based on this information, he changes his mind and elects to proceed with partial nephrectomy.

Figure 1.

Fagan nomogram (http://www.cebm.net) showing the application of the negative LR to determine the post-test probability of malignancy after a benign biopsy in a 55-year-old man with a 3-cm enhancing mass on CT.

Conclusion

You now appreciate that percutaneous biopsy of a small renal mass greatly influences the probability of malignancy. You are also reminded that biopsies are not perfectly accurate given subjectivity of histological interpretation and the occurrence of hybrid tumours [4]. Thus, even in the setting of a benign biopsy, patients should be carefully counselled, and if the patient chooses surveillance, serial imaging should be considered. You conclude that current data on the accuracy of renal mass biopsy is based on relatively weak evidence and studies from individual institutions may not be generalizable to other medical centres. You look forward to future multi-institutional studies with standardised biopsy techniques that will better define the performance of this diagnostic test.

Conflict of Interest

None declared.

Appendix: Appendix: Calculation of Negative LR for Percutaneous Biopsy of a Small Renal Mass and RCC

display math

Ancillary