ANATOMIC PATHOLOGY

Diagnosis of Chronic Endometritis in Biopsies with Stromal Breakdown

Hannah Gilmore, Deborah Fleischhacker and Jonathan L. Hecht

Plasma cells are the hallmark of chronic endometritis but are not specific for upper tract infection. Plasma cells have also been noted in hormonally mediated endometrial disorders in association with gland architectural changes (“disordered proliferative” and “anovulatory” patterns), and stromal breakdown. We reviewed benign endometrial biopsies diagnosed at Beth Israel Deaconess Medical Center over a 2-year period described as disordered/anovulatory patterns ± stromal breakdown. Cases were excluded if tissue was not available; women were younger than 50 years where most diagnoses were atrophic or cancer; or diagnoses were secretory, menstrual endometrium, or polyps. The remaining 61 cases were compared to 33 samples of unremarkable proliferative endometrium. Plasma cells were quantified on hematoxylin and eosin–stained sections and using a histochemical stain methyl green pyronin. The indication for biopsy was an abnormal pattern of bleeding in 34 cases, infertility workup in 7, incidental part of workup for pain, or other findings in 5. The majority of disordered proliferative endometrium had plasma cells (61% grade 1, 17% grade 2) all seen on methyl green pyronin staining only. Two thirds of proliferative endometrium with breakdown showed plasma cells (19% grade 1, 39% grade 2, 10 % grade 3). Plasma cells were rare in inactive endometrium and noted in only 18% of unremarkable proliferative endometrium, all grade 1. Plasma cells are commonly present in the endometrium of women with dysfunctional uterine bleeding and focal stromal breakdown. Given the lack of clinical evidence for infection, the inflammation likely represents a physiologic process.

Human Pathology, Article in Press, doi:10.1016/j.humpath.2006.09.002  

 

Histopathology of Serrated Adenoma, Its Variants, and Differentiation From Conventional Adenomatous and Hyperplastic Polyps

Shuan C. Li, Lawrence Burgart

Context.—Serrated adenomas can be morphologically subdivided into traditional and sessile types. They are thought to have a comparable rate of cancer progression like conventional adenomas, but they potentially have a faster rate of growth through methylation pathway(s). They share similar morphologic features with both the conventional adenoma and the hyperplastic polyp in a fashion that is different from a mixed adenoma and a hyperplastic polyp.

Objective.—To describe the histopathologic features of traditional serrated adenoma and sessile serrated adenoma and their comparison with traditional adenomas and hyperplastic polyp.

Data Sources.—Relevant articles in peer-review journals and the authors' working experience as practicing surgical pathologists with a specific interest in gastrointestinal pathology.

Conclusions.—Both types of serrated adenomas, traditional serrated adenoma and sessile serrated adenoma, are morphologically distinct, clinically important entities, and they can be diagnosed accurately in routine practice.

Epithelial polyps of the colorectum have traditionally been classified into two major subgroups with regard to their malignant potential: hyperplastic polyp (HPP) and adenoma. Adenomas were subdivided morphologically into several subgroups based on the degree of villous architecture as tubular, villous, or tubulovillous types (TA, VA, or TVA, respectively). In addition, there are a number of less-common polyps that are associated with specific clinical conditions, such as juvenile retention polyps, hamartomatous polyps, or inflammatory polyps. There was a consensus through the mid-1990s that adenomas, particularly those with villous architecture, were the precursors to colorectal carcinoma, and HPPs were not. This hypothesis was supported by studies on a rare familial adenomatous polyposis syndrome and sporadic colon cancers, which led to the recognition of APC/β-catenin pathway being the dominant molecular process involved in the adenoma-carcinoma sequence.

COMMENT

Colorectal adenocarcinoma can be subdivided into a suppressor phenotype (also known as microsatellite stable carcinoma) and a mutator phenotype (also known as MSI). Classically, microsatellite stable carcinomas were understood to be derived from progression and transformation of conventional adenomas through biallelic APC inactivation and abrogation of the Wnt/β-catenin pathway, as first described by Vogelstein et al. By contrast, MSI tumors, carcinoma, or serrated polyps have neither APC mutations nor elevated nuclear β-catenin levels. In fact, these tumors demonstrate BRAF mutation and hypermethylation of CpG island of hMLH-1 mismatch repair complex (serrated pathway). Studies have shown that the development of carcinoma via the serrated pathway might be more rapid than via the APC (suppressor) pathway. Because of their different genetic pathways, and probably different clinical course, it is important to attempt to recognize these similar but distinguishable types of polyps.

The diagnosis of SSA from a practical standpoint is more difficult compared with that of traditional SA (TSA), in that SSA does not have overt cytologic dysplasia, and it therefore resembles HPP. This leads to misdiagnosis of SSA as HPP, an underdiagnosis from the perspective of malignant risk. HPPs over 0.5 cm are probably rare proximal to the splenic flexure (or even in the descending colon); therefore, size and location can serve as practical aids to the diagnosis of SSA in routine practice. Questions remain regarding the premalignant potential of polyps in the distal colorectum with the features of SSA, and of small SSAs in the proximal colon.

Authors have summarized the detailed morphologic features of TSA and SSA in comparison to HPP and traditional adenomas, with emphasis on the morphologic differential diagnosis among these similar but distinct colonic mucosal lesions. Authors believe these lesions, including the more recently described SSA, can be diagnosed with certainty by practicing surgical pathologists, and that such diagnosis is important to appropriate patient management.

Archives of Pathology and Laboratory Medicine: Vol. 131, No. 3, pp. 440–445.

 

Vulvar Varices: An Uncommon Entity in Surgical Pathology

Bell, Diana, Kane, Philip B., Liang, Sharon et al

Varicose veins in the vulvar and perivulvar area are seen in 4% of women. Most of them are secondary to pregnancy and usually regress spontaneously. Vulvar varicose veins are rare in nonpregnant women. When present, they can be seen alone, associated with leg varices or associated with venous malformations of the labia, clitoral area, or vagina with or without arteriovenous malformations on the limbs or trunk (Klippel-Trenaunay-Weber syndrome and Parkes-Weber syndrome). In some cases, vulvar varices are seen as part of the so-called "pelvic congestion syndrome." Clinically, vulvar varices may present as small isolated protrusions, mainly in the labia majora, or as large masses, involving the vulva and even the perivulvar area. The treatment of choice of vulvar varices seen during pregnancy is conservative and symptomatic. Surgical pathologists need to be aware of the existence of vulvar varicose veins and its possible presence in biopsy specimens. Vulvar varicose veins can be misdiagnosed clinically as cysts or masses mainly in the Bartholin gland area. Correct diagnosis of the lesion is important to determine appropriate therapy and to recognize the possibility of associated anatomical or pathological problems.

DISCUSSION

Varicose veins of the vulva are most commonly seen during pregnancy. They have been described in 10% to 20% of pregnant women. They usually appear during the third or fourth month of gestation and regress spontaneously after delivery. In some cases, the varices may persist postpartum and may be symptomatic. The symptoms are usually perimenstrual and include tension, pain, and swelling of the vulva and the veins in the area.

Vulvar varicose veins are rare in nonpregnant women. When present, they can be seen alone, associated with leg varices or associated with venous malformations of the labia, clitoral area, or vagina with or without arteriovenous malformations on the limbs or trunk (Klippel-Trenaunay-Weber syndrome and Parkes-Weber syndrome). In some cases, vulvar varices are seen as part of the so-called "pelvic congestion syndrome." Pelvic congestion syndrome consists of pelvic pain, dyspareunia, dysmenorrhea, dysuria, and vulvar and perivulvar varices. It is seen in women with history of 2 or more pregnancies, and it is secondary to venous insufficiency of either the ovarian veins, internal iliac vein, or gonadal veins.

Clinically, vulvar varices may present as small isolated protrusions, mainly in the labia majora, or as large masses, involving the vulva and even perivulvar area.

The treatment of choice of vulvar varices seen during pregnancy is conservative and symptomatic (firm support and application of pressure). If the symptoms persist for more than 3 months after delivery, the varices can be treated with sclerotherapy. In patients with pelvic congestion syndrome, surgical ligation of the incompetent veins is the treatment of choice, although in some mild cases successes can be achieved with either local excision of the vulvar varices or sclerotherapy. Ligation of the incompetent veins can be done laparoscopically in most cases.

The 3 cases in our report are interesting because none of them presented clinically as typical varicose veins. Two of them were considered to be vulvar/Bartholin cysts and one an enlarged Bartholin gland. Two patients were asymptomatic (patients 1 and 2). The third patient complained of local pain and discomfort. In the asymptomatic patients, the lesions that were excised were small and measured 0.7 cm each in largest dimension. In contrast, in case 3, the specimen was larger (3.2 cm) and contained multiple dilated veins that were not only thrombosed but also were acutely inflamed, which correlates well with the clinical symptoms of pain and discomfort.

Patients 1 and 3 had been pregnant in the past. In patient 3, her pregnancy was one year before the diagnosis of her symptomatic vulvar varices. It is possible that, in this case, her vulvar varices were persistent lesions secondary to her pregnancy and that they became symptomatic because of thrombosis and superimposed inflammation (thrombophlebitis). The causes of this late event are uncertain.

In patients 2 and 3, the vulvar varices may have been residual changes secondary to previous pregnancy or may have been related to possible lower extremity varices.

None of the 3 patients had other anomalies as might be seen in any of the congenital malformation syndromes associated with vulvar varices (Klippel-Trenaunay-Weber syndrome and Parkes-Weber syndrome). None of them have symptoms of the pelvic congestion syndrome or any current gynecologic malignancy that may occasionally cause vulvar varices.

Surgical pathologists need to be aware of the existence of vulvar varicose veins and its possible presence in biopsy specimens. Based on this small series, vulvar varicose veins can be misdiagnosed clinically as cysts or masses mainly in the Bartholin gland area. Correct diagnosis of the lesion is important to determine appropriate therapy and to recognize the possibility of associated anatomical or pathological problems (congenital malformation in Klippel-Trenaunay-Weber syndrome, pelvic congestion syndrome, etc).

International Journal of Gynecological Pathology. 26(1):99-101, January 2007.

 

Clinicopathologic Features of Rhabdomyosarcoma of Gynecologic Origin in Adults

Ferguson, Sarah E., Gerald, William, Barakat, Richard R., Chi, Dennis S et al

Rhabdomyosarcoma (RMS) is the most common soft tissue tumor found in children. Up to 20% of RMS tumors in children originate in the genital tract making this the second most common site. RMS of gynecologic origin in adults is much less common. The purpose of this study was to describe the clinical and pathologic features of RMS of the adult female genital tract. We reviewed the histologic slides of women 16 years of age and older and included them in our study if they contained the classic histologic features of RMS as described by the 2002 World Health Organization classification of tumors. Rhabdomyoblastic components present in other established malignancies were not studied. Authors identified 15 patients, with a median age of 48 years (range, 16 to 69). Eleven (73%) of the tumors were of embryonal histology (cervix, 8; uterus, 2; and ovary, 1). Of the remaining 4 tumors, 2 were of alveolar (vulva) and 2 of pleomorphic (uterus, 1; fallopian tube, 1) histologic subtype. The majority (79%) of these patients presented with locoregional disease and had surgery as their primary intervention (73%). The median progression-free survival (PFS) and disease-specific survival was 9 months [95% confidence interval (CI), 1-24] and 21 months (95% CI, 14-28), respectively. The 5-year disease-specific survival was only 29%. There was a significant difference in PFS between cases of embryonal compared to cases with nonembryonal histology (19 vs. 3 mo, respectively) (P=0.04). Adult RMS of gynecologic origin presents with locoregional disease and most are morphologically similar to pediatric RMS; however, adult RMS behaves more aggressively, with worse overall survival. It is unclear whether these divergent outcomes are the result of differences in clinical management or because these tumors are biologically distinct.

 

American Journal of Surgical Pathology. 31(3):382-389, March 2007.

 

 

Immunohistochemical Marker Panels for Distinguishing between Epithelioid Mesothelioma and Lung Adenocarcinoma

Kushitani K, Takeshima Y, Amatya VJ et al

The distinction between epithelioid mesothelioma and lung adenocarcinoma remains an important diagnostic challenge for surgical pathologists. The aim of the present study was to select a limited and appropriate panel of antibodies that can differentiate between epithelioid mesothelioma and lung adenocarcinoma. Specimens of 90 epithelioid mesotheliomas and 51 lung adenocarcinomas obtained from Japanese cases were examined using calretinin, WT1, AE1/AE3, CAM5.2, cytokeratin (CK) 5/6, vimentin, epithelial membrane antigen (EMA), thrombomodulin, CEA, CA19-9, and CA125. Ninety-six percent of epithelioid mesotheliomas were positive for calretinin; 99% for WT1; 100% for AE1/AE; 97% for CAM5.2; 70% for CK 5/6; 91% for vimentin; 96% for EMA; 71% for thrombomodulin; 77% for mesothelin; 7% for CEA; 17% for CA19-9; and 85% for CA125. In contrast, 33% of lung adenocarcinomas were positive for calretinin; 16% for WT1; 100% for AE1/AE3, CAM5.2, and EMA; 41% for CK 5/6; 47% for vimentin; 20% for thrombomodulin; 69% for mesothelin; 98% for CEA; 73% for CA19-9; and 80% for CA125. For distinguishing between epithelioid mesothelioma and lung adenocarcinoma, the combination of CEA, calretinin and each WT1 or thrombomodulin was suggested to be the best panel of immunohistochemical markers.

Pathol Int. 2007 Apr;57(4):190-9.

  

Liver Histology in Patients with Sporadic Acute Hepatitis E: A Study of 11 Patients from South-West France

Peron JM, Danjoux M, Kamar N, Missoury R, Poirson H, Vinel JP, Mansuy JM, Bureau C, Izopet J, Brousset P, Selves J.

Hepatitis E virus is a ribonucleic acid (RNA) enterically transmitted virus that causes both epidemics and sporadic cases of acute hepatitis E in many countries of Asia and Africa. Domestically acquired (non-travel-associated) hepatitis E has been reported recently in many industrialized countries including the USA, Europe, and Japan. There is little information available on liver histology in these patients. Authors report a series of 11 patients with sporadic acute hepatitis E and needle liver histology in South-West France. Hepatitis E was diagnosed based on elevated transaminases (>10 upper limit normal) and the presence of specific serum antibodies (immunoglobulin-G class, present in all 11 patients) and/or viral RNA detection in serum and/or stools. Acute hepatitis lesions were observed in all cases with marked necro-inflammatory activity in nine patients. Confluent necrosis was present in five cases. Anisocaryosis and Kupffer's cell aggregates with siderosis were observed in most of the 11 patients. Cholangitis was frequent (9/11 cases). Cholestasis was observed in eight cases. Pseudo-glandular pattern was present in only one case but without zonal repartition. Characteristic pathological signs of acute hepatitis E were severe intralobular necrosis, polymorph inflammation, and acute cholangitis with numerous neutrophils.

 Virchows Arch. 2007 Feb 28; [Epub ahead of print]

 

CLINICAL PATHOLOGY

Novel Biomarker of Cardiac Mortality Risk Identified

In Cardiovascular News, 21 February 2007

A low blood level of the protein RANTES is an independent predictor of cardiac death in patients with coronary artery disease, study findings indicate.

Researchers found that men undergoing coronary angiography who had baseline levels of RANTES (Regulated upon Activation, Normal T-cell Expressed, and Secreted) in the lowest tertile had significantly higher cardiac mortality than those with RANTES levels in the higher two tertiles.

Researchers led by Erdal Cavusoglu, from the SUNY Downstate Medical Center in New York, USA, measured RANTES levels in 389 men undergoing coronary angiography for a variety of indications.

Over 2 years of follow-up there were 51 (13%) deaths, of which 31 (61%) were classified as related to cardiac etiology, the primary endpoint of the study.

Using the lowest tertile of RANTES values (<1949.80 pg/ml) as a cut-off, the primary end point of cardiac death at 24 months was 12.7% among those with low levels of RANTES compared with 6.0% among those with higher levels (p=0.029).

The secondary endpoint of MI at 24 months occurred in 61 (16%) patients, 10 of which were fatal.

Further analysis of the whole cohort showed that RANTES was an independent predictor of cardiac mortality (hazard ratio [HR]=0.64, p=0.0165). However, it was not a predictor of MI.

The researchers also analyzed the data according to whether or not patients had acute coronary syndrome (ACS), defined as ST-elevation MI, non-ST-elevation MI, or unstable angina. This indicated that RANTES was also an independent predictor of cardiac mortality in those patients classed as non-ACS. Moreover, it was also an independent predictor of MI in this group.  

In addition, RANTES proved to be an independent predictor of cardiac mortality in the subset of patients with diabetes.

In contrast, RANTES did not independently predict either outcome in the ACS group.

 

"Importantly, this study is consistent with the recent observations that low, rather than high, RANTES levels are associated with the presence of established CAD," the authors conclude in the journal Arteriosclerosis, Thrombosis, and Vascular Biology.

They believe that upregulation of the RANTES receptor, which is known to have proatherosclerotic effects, or greater deposition of RANTES on atherosclerotic arteries such that circulating levels are diminished, may explain their observations.

Arterioscler Thromb Vasc Biol 2007; Advance online publication

 

Panel Agrees on Whom to Test, How to Treat Pre-Diabetes

In Diabetes Today, 27-February-2007

People who have pre-diabetes should undergo intensive lifestyle interventions, and possibly drug therapy, to reduce their risk of developing diabetes, as well as their long-term risk for developing diabetic complications, according to a consensus statement being published in the March issue of Diabetes Care.

A seven-member panel of experts convened by the American Diabetes Association last year developed these guidelines and others geared toward people who exhibit early metabolic abnormalities. The panel's report grew out of concerns arising from the growing epidemic of type 2 diabetes, which now affects more than 20 million Americans. The disease is expected to continue increasing dramatically worldwide over the next two decades.

Type 2 diabetes is frequently preceded by one of two conditions together thought of as "pre-diabetes." These conditions called impaired fasting glucose (IFG) and impaired glucose tolerance (IGT) are similar in that they represent a state of abnormal glucose regulation that is not yet high enough for a diagnosis of diabetes but is too high to be considered normal. While these two states may affect different groups of people, both ultimately lead to type 2 diabetes in the majority of cases. However, previous studies, including the Diabetes Prevention Program, have demonstrated that lifestyle interventions aimed at weight reduction and increased physical activity, and medications, can substantially reduce the development of diabetes.

The panel convened over a three-day period in 2006 to answer questions such as how IFG and IGT differ; whether they should be treated (and how); and who should be screened for these conditions. The answers to these and other questions are included in the 7-page consensus statement.

The report's recommendations include:

* Lifestyle interventions (losing 5-10 percent of body weight and moderate intensity physical activity for at least 30 minutes per day) for any person exhibiting IFG or IGT, to prevent/delay the onset of diabetes and to help reduce the long-term risk of developing diabetic complications.

* Making weight loss and obesity prevention priorities in the United States because of the strong association between obesity and type 2 diabetes. The panel advised intensive weight-loss counseling for those who need it; changes in school-based meals and exercise programs; community infrastructure changes that are conducive to frequent exercise; and legislation that promotes a healthy lifestyle.

* The use of metformin as optional drug therapy, limited to those with both IFG and IGT who also have one or more additional high risk factors, because it has been shown to be most effective for these populations.

* Screening for IFG/IGT for anyone who is at risk for diabetes.

 

PSA Test in Mid-Life Predicts Prostate Cancer in Old Age

Martha Kerr

NEW YORK (Reuters Health) Mar 06 - Results of a single total prostate-specific antigen (tPSA) test, done between the ages of 44 and 50, can reliably predict prostate cancer up to 20 years afterward, investigators at Memorial Sloan-Kettering Cancer Center in New York report.

Dr. Hans Lilja and colleagues analyzed the long-term predictive power of tPSA, free PSA (fPSA) and human kallikrein 2 (hK2) levels measured between the ages of 44 and 50 years in 21,277 men enrolled in the Malmo (Sweden) Preventive Project between 1974 and 1986. The analysis focused on 462 participants who later developed prostate cancer and 1222 matched controls who did not.

The risk of prostate cancer in older age ranged from 1.0% to 7.5% if the mid-life tPSA was 0.5 ng/mL or lower, the investigators report in the February 1st issue of the Journal of Clinical Oncology.

Men with a tPSA of 0.51 to 1.0 ng/ml had a 2.5-fold increase in prostate cancer risk compared with men with tPSA levels of 0.5 ng/ml and lower.

Men with tPSA levels of 2.0 to 3.0, which are often considered within the normal range, had a more than 19-fold increase in prostate cancer risk compared with men with tPSA levels of 0.5 ng/ml and lower.

Free PSA and hK2 levels at baseline were also significant predictors of prostate cancer.

"The current data suggest that early biochemical changes (ie, slightly increased release of PSA and hK2 into blood) indicate a predisposition to prostate cancer that may be detectable two decades before the disease is diagnosed clinically," Dr. Lilja and colleagues conclude.

They caution that "recommendations to undergo biopsy on the basis of our findings would be premature, given that biopsies performed 15 to 25 years before the cancer would otherwise be diagnosed may not be informative."

"Our data strongly suggest that we should encourage all men to get a PSA test at age 45 to 50, not to try to detect cancer, but to work out their long-term cancer risk," Dr. Lilja told Reuters Health.

"We should now focus our subsequent screening efforts on the men at highest risk, spending our time and energy making sure that these men come in for regular screening (e.g. yearly) and making sure that they get the very best tests available. Men at lower risk of cancer may not require such intensive screening."

J Clin Oncol 2007;25:431-436.

  

Prostate-Specific Membrane Antigen Expression as a Predictor of Prostate Cancer Progression

Perner S, Hofer MD, Kim R, Shah RB, Li H et al

Distinguishing aggressive prostate cancer from indolent disease represents an important clinical challenge, because current therapy may lead to over treatment of men with limited disease. The prostate-specific membrane antigen (PSMA) is a membrane-bound glycoprotein that is highly restricted to the prostate. Previously, studies analyzing the expression of PSMA have found an up-regulation in correlation with prostate cancer, particularly in advanced cancer. This association is ideal for an application as a prognostic marker. In the current study, we characterized PSMA expression in a high-risk cohort and evaluated its potential use as predictive marker of prostate-specific antigen (PSA) recurrence. PSMA expression was analyzed by immunohistochemistry using tissue microarrays composed of tumor samples from 450 patients. Protein intensity was recorded using a semiautomated quantitative microscope system (ACIS II; Clarient Chromavision Medical Systems, San Juan Capistrano, CA). PSMA expression levels differed significantly (P < .001) between benign prostatic tissue, localized prostate cancer, and lymph node metastases. Dividing the cohort into high- and low-PSMA expressing cancers based on the median area of positive staining, we found that high PSMA levels were associated with significant increase of PSA recurrence (P = .004). This was independent of clinical parameters such as lymph node tumor burden (lymph node density, >20%; P < .001), extraprostatic extension (P = .017), seminal vesicle invasion (P < .001), and high Gleason score (8-10, P = .006). In a multivariate model, PSMA expression and metastases to pelvic lymph nodes were significantly associated with time to PSA recurrence (HR, 1.4; 95% confidence interval, 1.1-2.8, P = .017; and hazard ratio, 5; 95% confidence interval, 2.6-9.7, P < .001, respectively). In summary, PSMA is independently associated with PSA recurrence in a high-risk cohort and thus might provide insight into the additional use of adjuvant therapy. Validation on other cohorts is required.

Hum Pathol. 2007 Feb 21; [Epub ahead of print]

 

BOTTOM LINE

Who’s Killing the Autopsy? A New Tool for Assessing the Causes of Falling Autopsy Rates

Roger W Byard

Death is a fearful thing.

William Shakespeare (Measure for measure)

It is well recognised that hospital autopsy rates have been declining for a number of years. It now seems to be almost an “accepted” fact that hospital autopsies are a procedure of the past, having been bypassed by newer and more effective technologies for evaluating illness and disease states.

This is, of course, completely incorrect. Every study into autopsy practice has demonstrated the usefulness and necessity of a careful dissection after death to determine not only the true cause of death, but also the response of the patient to various treatment modalities. The autopsy is, and will always be, the ultimate audit of clinical practice. Following appropriately conducted autopsies, it has been consistently demonstrated that suggested clinical causes of death may be inaccurate and require correcting; unsuspected but significant conditions are often discovered; and unexpected effects of therapy and procedures may be revealed. The autopsy provides an excellent basis for teaching students the fundamentals of anatomy and the manifestations of disease. It also provides important information on the effects of newer drugs on normal as well as on diseased tissues. Given these well established facts, the obvious question is, “Why are autopsy rates in such decline?”.

Pathologists often complain that the responsibility for the alarming fall in autopsy rates rests squarely on their clinical colleagues. Clinicians’ palpable lack of interest and their obvious reluctance to ask relatives to give consent for autopsy are seen in a very unfavourable and disapproving light in the corridors of pathology departments the world over. However, it must be recognised that the lighting in many pathology departments is suboptimal, often due to their temporary or basement locations, and that such perceptions may not always be absolutely correct.

For autopsies to be meaningful, there needs to be a clear understanding of the patients’ specific clinical issues. This means that contact and discussion between the treating clinicians and the supervising/autopsy pathologist needs to occur before, often during, and certainly immediately after autopsy examination. Accurate dissection with photographic and written documentation of abnormalities is essential, and reports need to be issued promptly, with presentation of findings immediately after the autopsy or at least at regular weekly autopsy rounds. Provisional and final reports, with clinicopathological correlations and clear gross and microscopic descriptions, are the responsibility of the pathologist. Failure to disseminate them in a timely fashion causes both the clinician and the pathologist to lose interest in the case. Once a month, a clinicopathological conference needs to be held for general hospital medical, nursing and ancillary staff and students to review interesting cases and to answer clinical questions.

Does any of this occur? I would suggest that it often does not, and that this lack of support of autopsy practice from within the heartland of pathology may be one of the real reasons for clinicians’ lack of interest and the fall in autopsy rates. It has been observed that hospital-based pathologists, excluding those involved in paediatric/perinatal or forensic practice, do not even seem to like autopsies. Autopsies are considered to be technically messy and physically demanding and to detract from the more “scientific” and sterile world of histological and molecular pathology. In fact, they are sometimes seen as some sort of atavistic medical throwback, to be undertaken only by second-rate academics and those who are unemployable in the real world of 21st century anatomical pathology! Proof of the lowly status of the autopsy can be seen in the delegation of responsibility within hospital departments. All too often, the most junior registrar with the least training in anatomical pathology is designated to perform autopsies. The autopsy may even be undertaken in the absence of a consultant. Often it is a mortuary technician — one with many years of experience but no formal training, except in the “university of life and the school of hard knocks” — who teaches the unhappy registrar autopsy techniques. Registrars complain about how difficult it is to get consultants to review cases once dissections have been completed; and yet the literature maintains, and surveys have found, that “the necropsy is an invaluable investigation which is currently under-used” and that “most pathologists consider autopsies to be valuable and important for quality assurance in health care”. However, perhaps we should judge pathologists by what they do, rather than what they are quoted as saying?

Could this be a scurrilous and unfair suggestion? Possibly. However, one way to assess the validity of these assertions would be to examine pathology department records to find out when the last time was that senior pathologists performed an autopsy examination by themselves. I would suggest that years might well run into decades. Perhaps the most impartial and unemotive way to assess this would be to use a scoring system to objectively determine a pathology department’s commitment to autopsy practice . . . And fortunately, one just happens to be available.

A short questionnaire to determine the status of the autopsy in your pathology department

1. When was the last time your head of department...

 

a. Performed an autopsy?

b. Observed an autopsy?

c. Touched a dead body?

d. Saw a dead body?

e. Read about a dead body?

 

(Scoring: 1–6 days ago, 5 points; 7–30 days ago, 4 points; 4–52 weeks ago, 2 points; 1–10 years ago, 0 points; > 1 decade ago, automatic fail)

 

2. Who teaches registrars autopsy techniques?

 

a. Pathologists 5 points

b. Other registrars 1 point

c. Mortuary technicians 2 points

d. What teaching? – 5 points

 

3. Do CPCs occur regularly?

 

a. Yes 5 points

b. No 0 points

c. Don’t understand the question – 5 points

 

4. What is the average time for provisional and final autopsy reports to reach clinicians?

 

a. 1–6 days 5 points

b. 7–30 days 3 points

c. 1–6 months 1 point

d. 6 months to 1 year 0 points

e. > 1 year – 5 points

f. After retirement automatic fail

 

5. When was the last time an attending clinician was seen in the autopsy room?

 

a. 1–6 days ago 5 points

b. 7–30 days ago 3 points

c. 1–6 months ago 1 point

d. 6 months to 1 year ago 0 points

e. > 1 year ago – 5 points

f. Not recognised as such automatic fail and escorted out by security

 

6. When was the last time senior pathologists waxed lyrical about an interesting autopsy case?

 

a. 1–6 days ago 5 points

b. 7–30 days ago 3 points

c. 1–6 months ago 1 point

d. 6 months to 1 year ago 0 points

e. > 1 year ago – 5 points

f. Never automatic fail

Interpretation of results. Frankly, if your score is anything less than 35 points, I would not be blaming clinicians for the fall in autopsy rates, but would be looking much closer to home.

CPC = clinicopathological conference.

MJA 2005; 183 (11/12): 654-655