ORIGINAL ARTICLE
The utility of routine polyp histopathology aer endoscopic sinus surgery David H. Yeh, MD1 , Jay Wong, MD2 , Stephanie Hoffbauer, BHSc (Hon)2 , Bret Wehrli, MD, FRCPC3 , Doron Sommer, MD, FRCSC2 and Brian W. Rotenberg, MD, MPH, FRCSC1
Background: Routine histopathological assessment is standard practice for nasal polyp specimens obtained during endoscopic sinus surgery (ESS) for chronic rhinosinusitis (CRS). Retrospective studies suggest that routine histopathology of nasal polyps shows few unexpected diagnoses that alter patient management. Our objective was to study the use of routine pathological analysis, and its cost to the healthcare system, in a prospective manner. Methods: A multicenter prospective assessment was performed from data collected between 2007 and 2013. Only cases of patients undergoing ESS for bilateral CRS were included. We excluded unilateral disease cases, and cases in which diagnoses other than polyps were suspected either preoperatively or intraoperatively. We then compared the preoperative diagnosis with the final histopathology and identified the rate of unexpected pathologies. A cost analysis was performed. Results: Only 4 of 866 pathological specimens were identified as having a clinically significant unexpected diagnosis. All unexpected pathologies in this series were benign. These 4 cases account for 0.46% of all specimens
C
hronic rhinosinusitis (CRS) is often quoted as affecting up to 5% of the population whereas nasal polyps are estimated to be present in 4% of the general population.1 Although medical therapy with antibiotics,
1 Department
of Otolaryngology–Head and Neck Surgery, Schulich School of Medicine and Dentistry, Western University, London, Ontario, Canada; 2 Department of Otolaryngology–Head and Neck Surgery, McMaster University, Hamilton, Ontario, Canada; 3 Department of Pathology, Schulich School of Medicine and Dentistry, Western University, London, Ontario, Canada Correspondence to: Brian W. Rotenberg, MD, MPH, FRCSC, St. Joseph’s Healthcare Centre, 268 Grosvenor St., London, Ontario, Canada N6A 4V2; e-mail: [email protected] Potential conflict of interest: None provided. Presented as a podium presentation at the ARS Section Meeting at the Combined Otolaryngology Spring Meetings (COSM) 2014, on May 14-18, 2014, Las Vegas, NV. Received: 11 March 2014; Revised: 30 May 2014; Accepted: 23 June 2014 DOI: 10.1002/alr.21378 View this article online at wileyonlinelibrary.com.
reviewed. This translates to a number needed to screen of 217 cases of bilateral CRS to discover 1 unexpected pathology. The associated cost for making an unexpected diagnosis was $19,192.73. Conclusion: Routine histopathology of nasal polyps in ESS for bilateral CRS with polyps yields few unexpected and management-altering diagnoses. It carries a significant cost to the healthcare system. In cases of bilateral CRS with no other concerning clinical features, clinicians should exercise judgment in submiing polyp specimens for pathology rather than routinely sending polyps for C 2014 ARS-AAOA, LLC. histopathologic analysis.
Key Words: economics; histopathology; nasal polyps; nose neoplasm; paranasal sinus neoplasm How to Cite this Article: Yeh DH, Wong J, Hoauer S, Wehrli B, Sommer D, Rotenberg BW. The utility of routine polyp histopathology aer endoscopic sinus surgery. Int Forum Allergy Rhinol. 2014;4:926–930.
steroids, and nasal saline rinses is often the mainstay of treatment, management algorithms recommend surgical removal of polyps in cases that are insufficiently responsive to medical treatment.1 As with most surgical procedures, it is considered routine to perform histopathologic analysis on specimens obtained during surgery.2–7 Although nasal polyps are a common finding in CRS, neoplasms and premalignant lesions can at times have intranasal findings that mimic nasal polyposis. The potential adverse health implications and medicolegal ramifications for missing a sinister diagnosis drive the argument that all nasal polyp specimens should be analyzed under the microscope.2, 8 The necessity for routine histopathology in cases of CRS with nasal polyps has been questioned in the literature. Some authors, using retrospective data, have argued that neoplasms and premalignant lesions can be suspected preoperatively by their unilaterality or by their suspicious or unusual appearance endoscopically, and that, moreover, these are very rare.2, 3, 9 The ever-increasing burden of
International Forum of Allergy & Rhinology, Vol. 4, No. 11, November 2014
926
Routine polyp histopathology is unnecessary
healthcare costs in the modern era further calls into question practices of questionable benefit, including the appropriateness of obtaining routine nasal polyp histopathology. Our study aims to elucidate the role of routine histopathology in a novel fashion by calculating the number needed to screen (NNS) to pick up 1 unexpected diagnosis during endoscopic sinus surgery (ESS) for routine CRS, and to determine the cost of routine histopathology.
Patients and methods The research ethics boards at Western University (London, Ontario, Canada) and McMaster University (Hamilton, Ontario, Canada) granted approval for this study (HSREB 102890). This study represents prospectively gathered data obtained from patients who underwent ESS with polypectomy at the 2 academic centers. At Western University, a single surgeon (B.W.R.) with subspecialty training in sinus surgery performed ESS for CRS with polyps between January 29, 2007, and August 20, 2012, at St. Joseph’s Health Care Centre in London, Ontario. At McMaster University, 3 surgeons based out of either Hamilton General Hospital or McMaster University Medical Centre performed ESS for patients with CRS with nasal polyposis between February 2007 and April 2013. Surgery was performed in cases where appropriate medical therapy had been attempted and failed. All patients included in this study had an initial consultation, endoscopic exam, and computed tomography (CT) of the sinuses. Inclusion criteria were as follows: (1) bilateral CRS where cases met American Academy of Otolaryngology (AAO) diagnostic criteria for CRS; and (2) presence of bilateral nasal polyps. Exclusion criteria were (1) unilateral disease, even if meeting AAO criteria; (2) previous histopathologic diagnosis of nasal malignancy or premalignancy; or (3) lesions clinically suspicious for a diagnosis other than polyposis (based on the clinicians’ judgment from endoscopic exam, CT imaging, or intraoperatively). Surgical specimens were fixed in formalin, embedded in paraffin, and serial microscopic slides were stained with hematoxylin-eosin. The site-specific pathologists reviewed the histopathologic specimens. Data was extracted from the consultation notes, CT reports, operative notes, postoperative progress notes, and final histopathology reports. The first 3 authors (D.H.Y., J.W., S.H.) reviewed the electronic charts from their respective centers to obtain patient age at the time of surgery, gender, clinical diagnosis, procedure type, and histopathologic diagnosis. Both centers collected data independently and the data was then combined and tabulated. The NNS was calculated by taking the number of functional ESS (FESS) cases in which histopathologic specimens were analyzed and dividing by the number of clinically significant unexpected histopathologic diagnoses.10 The real-time cost for routine analysis of nasal polyp specimens is not a straightforward calculation. It can depend on the method of remuneration provided by an institu-
927
International Forum of Allergy & Rhinology, Vol. 4, No. 11, November 2014
FIGURE 1. Chart review strategy.
tion to the pathologist and the variable cost in preparation of the slides. As such, to ensure that our cost calculation was not institutionally biased, we included 2 possible methodologies to calculate the cost of analyzing a bilateral polyp specimen. The first method is based off the estimated cost of analyzing nasal polyp specimens at Western University. The pathologist’s interpretation of a polyp specimen costs $48.65. In addition, the charge for the report preparation, which also includes the cost of preparing and staining the specimen, is an additional $40. This gives a total cost of $88.65 for each routine histopathologic analysis of a nasal polyp specimen. At McMaster University, the cost for the pathologists’ interpretation of the nasal polyp specimen was based off an estimate of the average time required to analyze bilateral polyp specimens and the pathologists’ salary per unit time. Most pathologists in Ontario are salaried and the average hourly wage is often quoted as $150. If analysis of bilateral polyp specimens requires approximately 5 minutes, then the overall cost from the pathologist is $12.50 (5 minutes × $2.50 per minute). Technical preparation of the specimen costs $16.54/block and 2 blocks are required for bilateral specimens giving a cost of $33.08 for specimen preparation. Additionally, the cost of the biopsy by the surgeon is $50.90 for bilateral specimens. The overall estimated cost for the biopsy, specimen preparation, and histopathologic analysis would be $96.48.
Results We reviewed 1336 cases of endoscopic sinus surgery at the 2 respective sites. Two hundred cases were excluded for unilateral disease and an additional 270 cases of bilateral ESS were excluded based on study criteria. Of these 270 cases, 78 cases had no pathology report and in an additional 33 cases, surgery was being performed to biopsy an undiagnosed sinonasal mass. In the remaining 159 cases, there was a preoperative diagnosis other than nasal polyposis (Table 1). Therefore, in total, 866 cases met our inclusion and exclusion criteria for further study. Of these only 4 cases yielded an unexpected pathologic finding (Fig. 1). There were 3 inverted papillomas, and 1 squamous papilloma (Table 2). Consequently, out of the 866 cases of bilateral ESS, routine histopathologic analysis identified 4
Yeh et al.
TABLE 3. Clinically relevant UEPs in ESS for bilateral nasal
TABLE 1. Summary of excluded cases
polyposis Reason for exclusion/preoperative diagnosis
Cases (n)
No pathology
78
CRS without nasal polyps
62
Study
UEPs/total patients (n/N)
%
4/866
0.46
0/85
0.0
Garavello and Gaini (2005)
8/2147
0.37
Romashko and Stankiewicz3 (2005)
0/277
0.0
12/3375
0.36
This study 5
Mucocele
37
Yaman et al. (2011) 8
Unknown sinonasal mass
33
Inverted papilloma
24
Combined data
Mycetoma
8
Nasal deformity
5
Nasal cysts
4
Sinonasal tumor
4
Rhinolith
3
Capillary hemangioma
2
Discussion
Hamartoma
2
Nasal adhesion
2
Osteoma
2
Squamous papilloma
2
Chordoma
1
Fibrous dysplasia
1
Controversy exists in the literature regarding the use of routine histopathology for bilateral nasal polyps. Although it has been well established that the incidence of unexpected pathologies is low in patients with bilateral CRS with nasal polyposis, there are significant health implications for a patient if an occult malignancy is missed. Moreover, the medicolegal ramifications of overlooking such a diagnosis favors routine histopathologic analysis. In contrast, because healthcare systems have finite resources, dedicating finances and manpower into routine histopathology for nasal polyps is a very low-yield investment. It is a generally accepted convention that unilateral lesions and any lesions that have an unusual or suspicious appearance should be biopsied and sent for histopathologic analysis.3, 5, 7–9 Our series, however, looked specifically at cases of CRS in which there was bilateral polyposis with no other diagnosis suspected. Our novel data demonstrated an extremely low incidence of unexpected findings. There are a number of retrospective studies in the literature that previously examined the rate of unexpected diagnoses in polypectomy specimens. Based on studies that included polyp specimens from cases of both unilateral and bilateral disease, the incidence of unexpected pathology ranges from 0% to 0.92%.2–5, 7, 8 In studies isolated to bilateral nasal polyposis, thus more comparable to ours, the pick-up rate of clinically significant unexpected diagnoses determined by histopathologic analysis varies from 0% to 0.37% (Table 3). Romashko and Stankiewicz3 reported on 868 cases of polypectomy, in which a subset of 277 cases presented with bilateral polyposis. Of these 277 cases, there was no unexpected pathology discovered. Garavello and Gaini8 reviewed 2147 cases of bilateral polyposis in which they discovered 8 clinically relevant unexpected pathologies: 7 inverted papillomas and 1 case of adenocarcinoma. More recently, Yaman et al.5 analyzed 85 cases of bilateral polyposis and found no unexpected pathologies. In total, 1 case of occult malignancy
CRS = chronic rhinosinusitis.
TABLE 2. Clinically significant UEPs UEP
n
Inverted papilloma
3
Squamous papilloma
1
UEP = unexpected pathology.
(0.46%) clinically significant unexpected diagnoses. The remaining 823 histopathologies (99.54%) were interpreted as acute inflammation, chronic inflammation, nasal polyps, normal sinonasal mucosa, allergic mucin, or allergic fungal sinusitis. Based on our data, the pick-up rate in performing routine histopathology of nasal polyps is 0.46% (4/866). Therefore, the NNS is 216.5 cases of CRS with nasal polyps in order to identify a single unexpected clinically significant pathological finding. The estimated cost for routine histopathology for bilateral nasal polyps was calculated using the 2 different methodologies. At Western University, the cost was determined as $88.65 for each routine histopathologic analysis of a nasal polyp specimen. The cost of routine screening to identify a single clinically significant unexpected diagnosis would then be calculated as $19,192.73 (216.5 × $88.65).10 At McMaster University, the estimated cost for the routine histopathologic analysis of bilateral nasal specimens was $96.48. Using this value, we would expect rou-
ESS = endoscopic sinus surgery; UEP = unexpected pathology.
tine screening to identify 1 clinically significant diagnosis to cost $20,887.92 (216.5 × $96.48).10
International Forum of Allergy & Rhinology, Vol. 4, No. 11, November 2014
928
Routine polyp histopathology is unnecessary
was discovered among the 3375 cases of bilateral CRS with polyps (0.029%). In terms of unexpected pathologies, 12 clinically relevant unexpected pathologies were discovered after reviewing 3375 cases of CRS with bilateral polyps (0.36%). This represents an NNS of 281 cases (3375/12) to discover 1 clinically relevant histopathologic diagnosis. All of the 3 retrospective studies recommended against routine histopathology in cases of bilateral nasal polyposis.3, 5, 8 Although the proposal set forth in this article counters the prevailing surgical culture wherein a pathologist analyzes all specimens, there are an increasing number of surgeons who are questioning the need for routine histopathology in cases of benign disease. Another example of this within the field of rhinology is the routine pathological examination of septoplasty specimens. In their retrospective review of 303 cases of septoplasty, Haffey et al.11 demonstrated that although pathologists examine the cartilage specimens, no clinically significant unexpected diagnosis was discovered. They concluded that routine pathology of septoplasty specimens provided no clinical utility and burdened hospitals and patients alike with direct and indirect costs.11 Similarly, Randall et al.12 demonstrated through a review of the literature that routine histopathologic analysis of tonsil and adenoid specimens yielded an extremely low rate of occult malignancy (0.011% in nearly 55,000 cases).12 The authors also estimated the potential cost savings of selective tonsil and adenoid pathology at just over $35 million per year in the United States.12 These aforementioned studies have challenged the paradigm of routine surgical histopathology. In particular, the low rate of occult malignancy as demonstrated by Randall et al.12 has been established as a benchmark against which other studies are compared when considering discontinuation of routine pathology. Although the pick-up rate of clinically significant unexpected polyp pathologies in this study was higher than the rate of occult malignancy with routine tonsil and adenoid pathology, it should be noted that our calculations included nonmalignant tumors. In fact, only a single malignant pathology (adenocarcinoma) was unexpectedly identified with routine histopathology. An additional 10 cases of inverted papilloma and 1 case of squamous papilloma were uncovered. These tumors have an associated malignancy rate of approximately 10%.13 Therefore, even though this study endorses a pick-up rate of 0.36%, the rate of occult malignancy with bilateral polyposis is likely to be much lower. Aside from revealing an unexpected sinonasal malignancy, detailed polyp histopathology can also offer insightful clinical information about the patients’ disease process. For instance, if histopathology reveals a predominance of fungus and eosinophils vs neutrophils, postoperative medical treatment may be altered to better suit the patients’ conditions.14, 15 Furthermore, detailed histopathology and immunohistochemistry of polyps obtained routinely during surgery can be used for research
929
International Forum of Allergy & Rhinology, Vol. 4, No. 11, November 2014
to better understand the pathophysiology of CRS with polyposis. From a financial perspective, our series showed that the cost of discovering a single unexpected pathology is between $19,192.73 and $20,887.92. All unexpected pathologies in our study were benign lesions (papillomas). Malignancy is associated with inverted papillomas in roughly 10% of cases.13 We could therefore estimate the cost of discovering a single unexpected malignant pathology to be 10 times the cost of identifying a single unexpected pathology, or between $191,927.30 and $208,879.20. When we consider this in the context of the Canadian healthcare system where the average expenditure per capita was $5948 in 2012, routine histopathology of polyps bears a very significant and increasingly unjustifiable cost.16 Our study has inherent limitations. One of the major limitations is the possibility of sampling error. When ESS is performed for nasal polyps, a representative sample is taken for histopathologic analysis. There is the possibility that occult pathology could be left behind or removed without being analyzed during the course of surgery, resulting in a false negative. In some centers, specimen traps are used to collect all tissue samples removed from the sinonasal cavity. Although this might theoretically lower the false-negative rate, in actuality, it would not be feasible to analyze the entire pathologic specimen under the microscope. Furthermore, from a clinical point of view, using a specimen trap makes it impossible to direct treatment without knowing from which site specimens were obtained. Another potential limitation to our study is its external validity. All patients in this study were seen in tertiary care academic hospitals where the diagnosis was often premade and patients may have already had previous sinus surgery, as opposed to a community hospital where patients would be less differentiated and present without previous pathology reports. Last, although our study does examine the ad hoc cost of routine histopathology, it does not quantify the potential loss of productivity or the medicolegal costs associated with a missed or delayed diagnosis of a sinonasal malignancy. These could theoretically represent substantial costs—this issue may be worthy of a more formal cost analysis.
Conclusion The incidence of unexpected pathological findings for routine cases of patients with CRS with polyps undergoing ESS is extremely low. Our study shows an NNS of 216.5, and when pooled with data from other studies the NNS rises to 281. These data suggest that routine histopathologic examination in cases of bilateral nasal polyposis without any other suspicious findings is unwarranted. Furthermore, the substantial cost incurred by routine analysis further emphasizes the need to be judicious with biopsies. In cases of unilateral disease, or in cases where there is
Yeh et al.
any unusual feature seen endoscopically, histopathology is certainly warranted. However, the evidence from our study and from the scientific literature no longer supports routine polyp pathological analysis in cases of CRS. The evidence
strongly suggests that in the setting of bilateral polyposis, surgeons should exercise clinical judgment in requesting histopathologic analysis of polyps instead of arranging for it routinely.
References 1.
2.
3.
4.
5.
Desrosiers M, Evans G, Keith P, et al. Canadian Clinical Practice Guidelines for Acute and Chronic Rhinosinusitis. J Otolaryngol Head Neck Surg. 2011;40(Suppl 2):S99-S193. ˜ Kale SU, Mohite AU, Rowlands D, Drake-Lee AB. Clinical and histopathological correlation of nasal polyps: are there any surprises? Clin Otolaryngol. 2001;26:321–323. Romashko AA, Stankiewicz JA. Routine histopathology in uncomplicated sinus surgery: is it necessary? Otolaryngol Head Neck Surg. 2005;132:407–412; discussion 413. Van den Boer C, Brutel G, de Vries N. Is routine histopathological examination of FESS material useful? Eur Arch Otorhinolaryngol. 2010;267:381–384. Yaman H, Alkan N, Yilmaz S, Koc S, Belada A. Is routine histopathological analysis of nasal polyposis specimens necessary? Eur Arch Otorhinolaryngol. 2011;268:1013–1015.
6.
Younis RT, Hesse SV, Anand VK. Evaluation of the utility and cost-effectiveness of obtaining histopathologic diagnosis on all routine tonsillectomy specimens. Laryngoscope. 2001;111:2166–2169. 7. Diamantopoulos I, Jones N, Lowe J. All nasal polyps need histological examination: an auditbased appraisal of clinical practice. J Laryngol Otol. 2001;115:352. 8. Garavello W, Gaini RM. Histopathology of routine nasal polypectomy specimens: a review of 2,147 cases. Laryngoscope. 2005;115:1866–1868. 9. Hosemann W, Draf C. Danger points, complications and medico-legal aspects in endoscopic sinus surgery. GMS Curr Top Otorhinolaryngol Head Neck Surg. 2013;12:Doc06. 10. Rembold CM. Number needed to screen: development of a statistic for disease screening. BMJ. 1998;317:307–312. 11. Haffey T, Pabon S, Hawley K, Hoschar A, Sindwani R. Exploring the clinical value and implications of rou-
12.
13.
14.
15. 16.
tine pathological examination of septoplasty specimens. Laryngoscope. 2012;122:2373–2377. Randall DA, Martin PJ, Thompson LD. Routine histologic examination is unnecessary for tonsillectomy or adenoidectomy. Laryngoscope. 2007;117:1600– 1604. Von Buchwald C, Bradley PJ. Risks of malignancy in inverted papilloma of the nose and paranasal sinuses. Curr Opin Otolaryngol Head Neck Surg. 2007;15:95–98. Chan Y, Kuhn FA. An update on the classifications, diagnosis, and treatment of rhinosinusitis. Curr Opin Otolaryngol Head Neck Surg. 2009;17:204–208. Hellquist HB. Histopathology. Allergy Asthma Proc. 1996;17:237–242. Canadian Institute for Health Information. Canada’s health care spending growth slows. October 30, 2012. http://www.cihi.ca/cihi-ext-portal/internet/en/ document/spending+and+health+workforce/spending/ release_30oct12#.
International Forum of Allergy & Rhinology, Vol. 4, No. 11, November 2014
930
The utility of routine polyp histopathology aer endoscopic sinus surgery David H. Yeh, MD1 , Jay Wong, MD2 , Stephanie Hoffbauer, BHSc (Hon)2 , Bret Wehrli, MD, FRCPC3 , Doron Sommer, MD, FRCSC2 and Brian W. Rotenberg, MD, MPH, FRCSC1
Background: Routine histopathological assessment is standard practice for nasal polyp specimens obtained during endoscopic sinus surgery (ESS) for chronic rhinosinusitis (CRS). Retrospective studies suggest that routine histopathology of nasal polyps shows few unexpected diagnoses that alter patient management. Our objective was to study the use of routine pathological analysis, and its cost to the healthcare system, in a prospective manner. Methods: A multicenter prospective assessment was performed from data collected between 2007 and 2013. Only cases of patients undergoing ESS for bilateral CRS were included. We excluded unilateral disease cases, and cases in which diagnoses other than polyps were suspected either preoperatively or intraoperatively. We then compared the preoperative diagnosis with the final histopathology and identified the rate of unexpected pathologies. A cost analysis was performed. Results: Only 4 of 866 pathological specimens were identified as having a clinically significant unexpected diagnosis. All unexpected pathologies in this series were benign. These 4 cases account for 0.46% of all specimens
C
hronic rhinosinusitis (CRS) is often quoted as affecting up to 5% of the population whereas nasal polyps are estimated to be present in 4% of the general population.1 Although medical therapy with antibiotics,
1 Department
of Otolaryngology–Head and Neck Surgery, Schulich School of Medicine and Dentistry, Western University, London, Ontario, Canada; 2 Department of Otolaryngology–Head and Neck Surgery, McMaster University, Hamilton, Ontario, Canada; 3 Department of Pathology, Schulich School of Medicine and Dentistry, Western University, London, Ontario, Canada Correspondence to: Brian W. Rotenberg, MD, MPH, FRCSC, St. Joseph’s Healthcare Centre, 268 Grosvenor St., London, Ontario, Canada N6A 4V2; e-mail: [email protected] Potential conflict of interest: None provided. Presented as a podium presentation at the ARS Section Meeting at the Combined Otolaryngology Spring Meetings (COSM) 2014, on May 14-18, 2014, Las Vegas, NV. Received: 11 March 2014; Revised: 30 May 2014; Accepted: 23 June 2014 DOI: 10.1002/alr.21378 View this article online at wileyonlinelibrary.com.
reviewed. This translates to a number needed to screen of 217 cases of bilateral CRS to discover 1 unexpected pathology. The associated cost for making an unexpected diagnosis was $19,192.73. Conclusion: Routine histopathology of nasal polyps in ESS for bilateral CRS with polyps yields few unexpected and management-altering diagnoses. It carries a significant cost to the healthcare system. In cases of bilateral CRS with no other concerning clinical features, clinicians should exercise judgment in submiing polyp specimens for pathology rather than routinely sending polyps for C 2014 ARS-AAOA, LLC. histopathologic analysis.
Key Words: economics; histopathology; nasal polyps; nose neoplasm; paranasal sinus neoplasm How to Cite this Article: Yeh DH, Wong J, Hoauer S, Wehrli B, Sommer D, Rotenberg BW. The utility of routine polyp histopathology aer endoscopic sinus surgery. Int Forum Allergy Rhinol. 2014;4:926–930.
steroids, and nasal saline rinses is often the mainstay of treatment, management algorithms recommend surgical removal of polyps in cases that are insufficiently responsive to medical treatment.1 As with most surgical procedures, it is considered routine to perform histopathologic analysis on specimens obtained during surgery.2–7 Although nasal polyps are a common finding in CRS, neoplasms and premalignant lesions can at times have intranasal findings that mimic nasal polyposis. The potential adverse health implications and medicolegal ramifications for missing a sinister diagnosis drive the argument that all nasal polyp specimens should be analyzed under the microscope.2, 8 The necessity for routine histopathology in cases of CRS with nasal polyps has been questioned in the literature. Some authors, using retrospective data, have argued that neoplasms and premalignant lesions can be suspected preoperatively by their unilaterality or by their suspicious or unusual appearance endoscopically, and that, moreover, these are very rare.2, 3, 9 The ever-increasing burden of
International Forum of Allergy & Rhinology, Vol. 4, No. 11, November 2014
926
Routine polyp histopathology is unnecessary
healthcare costs in the modern era further calls into question practices of questionable benefit, including the appropriateness of obtaining routine nasal polyp histopathology. Our study aims to elucidate the role of routine histopathology in a novel fashion by calculating the number needed to screen (NNS) to pick up 1 unexpected diagnosis during endoscopic sinus surgery (ESS) for routine CRS, and to determine the cost of routine histopathology.
Patients and methods The research ethics boards at Western University (London, Ontario, Canada) and McMaster University (Hamilton, Ontario, Canada) granted approval for this study (HSREB 102890). This study represents prospectively gathered data obtained from patients who underwent ESS with polypectomy at the 2 academic centers. At Western University, a single surgeon (B.W.R.) with subspecialty training in sinus surgery performed ESS for CRS with polyps between January 29, 2007, and August 20, 2012, at St. Joseph’s Health Care Centre in London, Ontario. At McMaster University, 3 surgeons based out of either Hamilton General Hospital or McMaster University Medical Centre performed ESS for patients with CRS with nasal polyposis between February 2007 and April 2013. Surgery was performed in cases where appropriate medical therapy had been attempted and failed. All patients included in this study had an initial consultation, endoscopic exam, and computed tomography (CT) of the sinuses. Inclusion criteria were as follows: (1) bilateral CRS where cases met American Academy of Otolaryngology (AAO) diagnostic criteria for CRS; and (2) presence of bilateral nasal polyps. Exclusion criteria were (1) unilateral disease, even if meeting AAO criteria; (2) previous histopathologic diagnosis of nasal malignancy or premalignancy; or (3) lesions clinically suspicious for a diagnosis other than polyposis (based on the clinicians’ judgment from endoscopic exam, CT imaging, or intraoperatively). Surgical specimens were fixed in formalin, embedded in paraffin, and serial microscopic slides were stained with hematoxylin-eosin. The site-specific pathologists reviewed the histopathologic specimens. Data was extracted from the consultation notes, CT reports, operative notes, postoperative progress notes, and final histopathology reports. The first 3 authors (D.H.Y., J.W., S.H.) reviewed the electronic charts from their respective centers to obtain patient age at the time of surgery, gender, clinical diagnosis, procedure type, and histopathologic diagnosis. Both centers collected data independently and the data was then combined and tabulated. The NNS was calculated by taking the number of functional ESS (FESS) cases in which histopathologic specimens were analyzed and dividing by the number of clinically significant unexpected histopathologic diagnoses.10 The real-time cost for routine analysis of nasal polyp specimens is not a straightforward calculation. It can depend on the method of remuneration provided by an institu-
927
International Forum of Allergy & Rhinology, Vol. 4, No. 11, November 2014
FIGURE 1. Chart review strategy.
tion to the pathologist and the variable cost in preparation of the slides. As such, to ensure that our cost calculation was not institutionally biased, we included 2 possible methodologies to calculate the cost of analyzing a bilateral polyp specimen. The first method is based off the estimated cost of analyzing nasal polyp specimens at Western University. The pathologist’s interpretation of a polyp specimen costs $48.65. In addition, the charge for the report preparation, which also includes the cost of preparing and staining the specimen, is an additional $40. This gives a total cost of $88.65 for each routine histopathologic analysis of a nasal polyp specimen. At McMaster University, the cost for the pathologists’ interpretation of the nasal polyp specimen was based off an estimate of the average time required to analyze bilateral polyp specimens and the pathologists’ salary per unit time. Most pathologists in Ontario are salaried and the average hourly wage is often quoted as $150. If analysis of bilateral polyp specimens requires approximately 5 minutes, then the overall cost from the pathologist is $12.50 (5 minutes × $2.50 per minute). Technical preparation of the specimen costs $16.54/block and 2 blocks are required for bilateral specimens giving a cost of $33.08 for specimen preparation. Additionally, the cost of the biopsy by the surgeon is $50.90 for bilateral specimens. The overall estimated cost for the biopsy, specimen preparation, and histopathologic analysis would be $96.48.
Results We reviewed 1336 cases of endoscopic sinus surgery at the 2 respective sites. Two hundred cases were excluded for unilateral disease and an additional 270 cases of bilateral ESS were excluded based on study criteria. Of these 270 cases, 78 cases had no pathology report and in an additional 33 cases, surgery was being performed to biopsy an undiagnosed sinonasal mass. In the remaining 159 cases, there was a preoperative diagnosis other than nasal polyposis (Table 1). Therefore, in total, 866 cases met our inclusion and exclusion criteria for further study. Of these only 4 cases yielded an unexpected pathologic finding (Fig. 1). There were 3 inverted papillomas, and 1 squamous papilloma (Table 2). Consequently, out of the 866 cases of bilateral ESS, routine histopathologic analysis identified 4
Yeh et al.
TABLE 3. Clinically relevant UEPs in ESS for bilateral nasal
TABLE 1. Summary of excluded cases
polyposis Reason for exclusion/preoperative diagnosis
Cases (n)
No pathology
78
CRS without nasal polyps
62
Study
UEPs/total patients (n/N)
%
4/866
0.46
0/85
0.0
Garavello and Gaini (2005)
8/2147
0.37
Romashko and Stankiewicz3 (2005)
0/277
0.0
12/3375
0.36
This study 5
Mucocele
37
Yaman et al. (2011) 8
Unknown sinonasal mass
33
Inverted papilloma
24
Combined data
Mycetoma
8
Nasal deformity
5
Nasal cysts
4
Sinonasal tumor
4
Rhinolith
3
Capillary hemangioma
2
Discussion
Hamartoma
2
Nasal adhesion
2
Osteoma
2
Squamous papilloma
2
Chordoma
1
Fibrous dysplasia
1
Controversy exists in the literature regarding the use of routine histopathology for bilateral nasal polyps. Although it has been well established that the incidence of unexpected pathologies is low in patients with bilateral CRS with nasal polyposis, there are significant health implications for a patient if an occult malignancy is missed. Moreover, the medicolegal ramifications of overlooking such a diagnosis favors routine histopathologic analysis. In contrast, because healthcare systems have finite resources, dedicating finances and manpower into routine histopathology for nasal polyps is a very low-yield investment. It is a generally accepted convention that unilateral lesions and any lesions that have an unusual or suspicious appearance should be biopsied and sent for histopathologic analysis.3, 5, 7–9 Our series, however, looked specifically at cases of CRS in which there was bilateral polyposis with no other diagnosis suspected. Our novel data demonstrated an extremely low incidence of unexpected findings. There are a number of retrospective studies in the literature that previously examined the rate of unexpected diagnoses in polypectomy specimens. Based on studies that included polyp specimens from cases of both unilateral and bilateral disease, the incidence of unexpected pathology ranges from 0% to 0.92%.2–5, 7, 8 In studies isolated to bilateral nasal polyposis, thus more comparable to ours, the pick-up rate of clinically significant unexpected diagnoses determined by histopathologic analysis varies from 0% to 0.37% (Table 3). Romashko and Stankiewicz3 reported on 868 cases of polypectomy, in which a subset of 277 cases presented with bilateral polyposis. Of these 277 cases, there was no unexpected pathology discovered. Garavello and Gaini8 reviewed 2147 cases of bilateral polyposis in which they discovered 8 clinically relevant unexpected pathologies: 7 inverted papillomas and 1 case of adenocarcinoma. More recently, Yaman et al.5 analyzed 85 cases of bilateral polyposis and found no unexpected pathologies. In total, 1 case of occult malignancy
CRS = chronic rhinosinusitis.
TABLE 2. Clinically significant UEPs UEP
n
Inverted papilloma
3
Squamous papilloma
1
UEP = unexpected pathology.
(0.46%) clinically significant unexpected diagnoses. The remaining 823 histopathologies (99.54%) were interpreted as acute inflammation, chronic inflammation, nasal polyps, normal sinonasal mucosa, allergic mucin, or allergic fungal sinusitis. Based on our data, the pick-up rate in performing routine histopathology of nasal polyps is 0.46% (4/866). Therefore, the NNS is 216.5 cases of CRS with nasal polyps in order to identify a single unexpected clinically significant pathological finding. The estimated cost for routine histopathology for bilateral nasal polyps was calculated using the 2 different methodologies. At Western University, the cost was determined as $88.65 for each routine histopathologic analysis of a nasal polyp specimen. The cost of routine screening to identify a single clinically significant unexpected diagnosis would then be calculated as $19,192.73 (216.5 × $88.65).10 At McMaster University, the estimated cost for the routine histopathologic analysis of bilateral nasal specimens was $96.48. Using this value, we would expect rou-
ESS = endoscopic sinus surgery; UEP = unexpected pathology.
tine screening to identify 1 clinically significant diagnosis to cost $20,887.92 (216.5 × $96.48).10
International Forum of Allergy & Rhinology, Vol. 4, No. 11, November 2014
928
Routine polyp histopathology is unnecessary
was discovered among the 3375 cases of bilateral CRS with polyps (0.029%). In terms of unexpected pathologies, 12 clinically relevant unexpected pathologies were discovered after reviewing 3375 cases of CRS with bilateral polyps (0.36%). This represents an NNS of 281 cases (3375/12) to discover 1 clinically relevant histopathologic diagnosis. All of the 3 retrospective studies recommended against routine histopathology in cases of bilateral nasal polyposis.3, 5, 8 Although the proposal set forth in this article counters the prevailing surgical culture wherein a pathologist analyzes all specimens, there are an increasing number of surgeons who are questioning the need for routine histopathology in cases of benign disease. Another example of this within the field of rhinology is the routine pathological examination of septoplasty specimens. In their retrospective review of 303 cases of septoplasty, Haffey et al.11 demonstrated that although pathologists examine the cartilage specimens, no clinically significant unexpected diagnosis was discovered. They concluded that routine pathology of septoplasty specimens provided no clinical utility and burdened hospitals and patients alike with direct and indirect costs.11 Similarly, Randall et al.12 demonstrated through a review of the literature that routine histopathologic analysis of tonsil and adenoid specimens yielded an extremely low rate of occult malignancy (0.011% in nearly 55,000 cases).12 The authors also estimated the potential cost savings of selective tonsil and adenoid pathology at just over $35 million per year in the United States.12 These aforementioned studies have challenged the paradigm of routine surgical histopathology. In particular, the low rate of occult malignancy as demonstrated by Randall et al.12 has been established as a benchmark against which other studies are compared when considering discontinuation of routine pathology. Although the pick-up rate of clinically significant unexpected polyp pathologies in this study was higher than the rate of occult malignancy with routine tonsil and adenoid pathology, it should be noted that our calculations included nonmalignant tumors. In fact, only a single malignant pathology (adenocarcinoma) was unexpectedly identified with routine histopathology. An additional 10 cases of inverted papilloma and 1 case of squamous papilloma were uncovered. These tumors have an associated malignancy rate of approximately 10%.13 Therefore, even though this study endorses a pick-up rate of 0.36%, the rate of occult malignancy with bilateral polyposis is likely to be much lower. Aside from revealing an unexpected sinonasal malignancy, detailed polyp histopathology can also offer insightful clinical information about the patients’ disease process. For instance, if histopathology reveals a predominance of fungus and eosinophils vs neutrophils, postoperative medical treatment may be altered to better suit the patients’ conditions.14, 15 Furthermore, detailed histopathology and immunohistochemistry of polyps obtained routinely during surgery can be used for research
929
International Forum of Allergy & Rhinology, Vol. 4, No. 11, November 2014
to better understand the pathophysiology of CRS with polyposis. From a financial perspective, our series showed that the cost of discovering a single unexpected pathology is between $19,192.73 and $20,887.92. All unexpected pathologies in our study were benign lesions (papillomas). Malignancy is associated with inverted papillomas in roughly 10% of cases.13 We could therefore estimate the cost of discovering a single unexpected malignant pathology to be 10 times the cost of identifying a single unexpected pathology, or between $191,927.30 and $208,879.20. When we consider this in the context of the Canadian healthcare system where the average expenditure per capita was $5948 in 2012, routine histopathology of polyps bears a very significant and increasingly unjustifiable cost.16 Our study has inherent limitations. One of the major limitations is the possibility of sampling error. When ESS is performed for nasal polyps, a representative sample is taken for histopathologic analysis. There is the possibility that occult pathology could be left behind or removed without being analyzed during the course of surgery, resulting in a false negative. In some centers, specimen traps are used to collect all tissue samples removed from the sinonasal cavity. Although this might theoretically lower the false-negative rate, in actuality, it would not be feasible to analyze the entire pathologic specimen under the microscope. Furthermore, from a clinical point of view, using a specimen trap makes it impossible to direct treatment without knowing from which site specimens were obtained. Another potential limitation to our study is its external validity. All patients in this study were seen in tertiary care academic hospitals where the diagnosis was often premade and patients may have already had previous sinus surgery, as opposed to a community hospital where patients would be less differentiated and present without previous pathology reports. Last, although our study does examine the ad hoc cost of routine histopathology, it does not quantify the potential loss of productivity or the medicolegal costs associated with a missed or delayed diagnosis of a sinonasal malignancy. These could theoretically represent substantial costs—this issue may be worthy of a more formal cost analysis.
Conclusion The incidence of unexpected pathological findings for routine cases of patients with CRS with polyps undergoing ESS is extremely low. Our study shows an NNS of 216.5, and when pooled with data from other studies the NNS rises to 281. These data suggest that routine histopathologic examination in cases of bilateral nasal polyposis without any other suspicious findings is unwarranted. Furthermore, the substantial cost incurred by routine analysis further emphasizes the need to be judicious with biopsies. In cases of unilateral disease, or in cases where there is
Yeh et al.
any unusual feature seen endoscopically, histopathology is certainly warranted. However, the evidence from our study and from the scientific literature no longer supports routine polyp pathological analysis in cases of CRS. The evidence
strongly suggests that in the setting of bilateral polyposis, surgeons should exercise clinical judgment in requesting histopathologic analysis of polyps instead of arranging for it routinely.
References 1.
2.
3.
4.
5.
Desrosiers M, Evans G, Keith P, et al. Canadian Clinical Practice Guidelines for Acute and Chronic Rhinosinusitis. J Otolaryngol Head Neck Surg. 2011;40(Suppl 2):S99-S193. ˜ Kale SU, Mohite AU, Rowlands D, Drake-Lee AB. Clinical and histopathological correlation of nasal polyps: are there any surprises? Clin Otolaryngol. 2001;26:321–323. Romashko AA, Stankiewicz JA. Routine histopathology in uncomplicated sinus surgery: is it necessary? Otolaryngol Head Neck Surg. 2005;132:407–412; discussion 413. Van den Boer C, Brutel G, de Vries N. Is routine histopathological examination of FESS material useful? Eur Arch Otorhinolaryngol. 2010;267:381–384. Yaman H, Alkan N, Yilmaz S, Koc S, Belada A. Is routine histopathological analysis of nasal polyposis specimens necessary? Eur Arch Otorhinolaryngol. 2011;268:1013–1015.
6.
Younis RT, Hesse SV, Anand VK. Evaluation of the utility and cost-effectiveness of obtaining histopathologic diagnosis on all routine tonsillectomy specimens. Laryngoscope. 2001;111:2166–2169. 7. Diamantopoulos I, Jones N, Lowe J. All nasal polyps need histological examination: an auditbased appraisal of clinical practice. J Laryngol Otol. 2001;115:352. 8. Garavello W, Gaini RM. Histopathology of routine nasal polypectomy specimens: a review of 2,147 cases. Laryngoscope. 2005;115:1866–1868. 9. Hosemann W, Draf C. Danger points, complications and medico-legal aspects in endoscopic sinus surgery. GMS Curr Top Otorhinolaryngol Head Neck Surg. 2013;12:Doc06. 10. Rembold CM. Number needed to screen: development of a statistic for disease screening. BMJ. 1998;317:307–312. 11. Haffey T, Pabon S, Hawley K, Hoschar A, Sindwani R. Exploring the clinical value and implications of rou-
12.
13.
14.
15. 16.
tine pathological examination of septoplasty specimens. Laryngoscope. 2012;122:2373–2377. Randall DA, Martin PJ, Thompson LD. Routine histologic examination is unnecessary for tonsillectomy or adenoidectomy. Laryngoscope. 2007;117:1600– 1604. Von Buchwald C, Bradley PJ. Risks of malignancy in inverted papilloma of the nose and paranasal sinuses. Curr Opin Otolaryngol Head Neck Surg. 2007;15:95–98. Chan Y, Kuhn FA. An update on the classifications, diagnosis, and treatment of rhinosinusitis. Curr Opin Otolaryngol Head Neck Surg. 2009;17:204–208. Hellquist HB. Histopathology. Allergy Asthma Proc. 1996;17:237–242. Canadian Institute for Health Information. Canada’s health care spending growth slows. October 30, 2012. http://www.cihi.ca/cihi-ext-portal/internet/en/ document/spending+and+health+workforce/spending/ release_30oct12#.
International Forum of Allergy & Rhinology, Vol. 4, No. 11, November 2014
930
