The autopsy rate gradually decreased during 1950–1999, and increased during the most recent decade (2000–2009). The diagnostic inaccuracy rate was continuously high during the 60 years.
To investigate disagreement between the pathological and clinical diagnosis during 60 years (1950–2009).
A 60-year retrospective study was carried out on the 4140 autopsy cases performed in Zhejiang University School of Medicine.
The highest number of cases was 1037 during 1960–1969, while the lowest was 102 during 1990–1999. During the 1999–2009 period, 978 cases were completed, which ranked second within the 60 years. The total clinical misdiagnosis rate was 46.38%, while the highest was 73.82% in 2000–2009. During the 60 years, the diseases associated with highest diagnostic inaccuracy rates were circulatory diseases (76.97%), cancer (60.99%), and brain diseases (54.48%). The invasive fungal infection rate was 1.84% of the 4140 cases, and the diagnostic inaccuracy rate for this condition reached as high as 86.10%. In the autopsied disease spectrum over the 60 years, the most common diseases were respiratory (1349, 32.58%), circulatory (495, 11.96%), and brain diseases (424, 10.24%).
Citation: Zhu K, Feng H, Xu Y, Mao Z, Zhang W, Chen J, et al. (2014) An Analysis of 60 Years of Autopsy Data from Zhejiang University in Hangzhou, China. PLoS ONE 9(11): e112500. https://doi.org/10.1371/journal.pone.0112500
Editor: Wan-Xi Yang, Zhejiang University, China
Received: June 10, 2014; Accepted: October 18, 2014; Published: November 26, 2014
Copyright: © 2014 Zhu et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The authors confirm that all data underlying the findings are fully available without restriction. All relevant data are within the paper and its Supporting Information files.
Funding: This work was supported by the Zhejiang Provincial Natural Science Foundation (2010R50049), by the Zhejiang Medical and Health Science and Technology Plan Project (WKJ2013-2-009) and by the Scientific Research Foundation for Returned Overseas Scholars (Ministry of Education of China 2013). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
In China, hospitals are graded, and the Ministry of Health requires that hospitals belonging to the III and II grades must attain autopsy rates of at least 15% and 10%, respectively. In reality, these percentages are not achieved in many hospitals. For example, the number of autopsies conducted between 1990 and 1999 in Hangzhou (population 6 million), was only 102, which translates into an average of only 10 per year. In a span of 10 years, hospitals in Hangzhou have had only between 1 and 3 autopsies, or even less. The number of autopsies conducted is disproportionate to the size and grading of the hospitals in Hangzhou. In addition, many hospitals do not conduct clinicopathological conferences, resulting in problems in the training of new physicians as well as the development of clinical medicine. Here, an analysis of 4140 autopsies in the past 60 years conducted at Zhejiang University School of Medicine in Hangzhou between 1950 and 2009 was carried out to focus attention on the field.
Cases was obtained concerning 4140 autopsies conducted between 1950 and 2009 at Zhejiang University School of Medicine (known as Zhejiang Medical University prior to 1999). Disease analysis and the categorization of pathological diagnoses were in accord with the WHO guidelines on “International Standards for Diseases and Injury”. The autopsy center identifies the death reason, at the same time, it is the teaching and scientific research unit, and will carry out related photography, video, teaching and research activities before and after the inspection process. The patient family members understand and support the above works. The study was reviewed and approved by Zhejiang University School of Medicine Ethics Committee before the study began. The written informed consent was given by the next of kin/caregiver in the case of children or the deceased, for their clinical records to be used in this study. The patient records/information was anonymized and de-identified prior to analysis.
The number of autopsy cases in the 60 years, the gender and age of the autopsied individuals, comparison of the autopsy and clinical diagnoses, and the distribution of autopsy-determined fatal diseases were determined. In the event of a list of possible clinical diagnoses, only the first three were taken into account for statistical purposes. In the event where any one item from clinical diagnosis overlapped with an item from the autopsy-determined diagnosis, the two diagnoses were said to concur, otherwise being taken as misdiagnosis. We summarized and compared our findings with autopsy reports from some of the other major medical centers in China.
Fetuses of 20 weeks to neonates 28 days after birth comprised the perinatal group, >28 days to 7 years the infant group, >7 to 18 years the child and adolescent group, and >18 years the adult group. Further categorization was done according to the location of the fatal pathological change. At the same time, cases involving tuberculosis were categorized under infectious disease. Other cases that did not fall into the above categories formed the last group (others).
A total of 4140 autopsy cases were conducted between 1950 and 2009. Of these, 2785 were male and 1355 female (ratio 2.06∶1). The largest number of cases (1037) occurred between 1960 and 1969, giving an average of >100 per year. The fewest (102) were recorded between 1990 and 1999, giving an average of only 10 per year. The number of cases conducted at Zhejiang University School of Medicine reflected the standard number of autopsies in Hangzhou (Table 1).
The overall misdiagnosis rate was 46.38%. According to the statistics, this rate was highest between 2000 and 2009, reaching 73.82%, and lowest between 1980 and 1989, falling to 30.63%. The misdiagnosis rate between 1990 and 1999 was 41.18%. During the 60 years, the highest misdiagnosis rate was for circulatory diseases (76.97%), followed by tumors (60.99%) and brain diseases (54.48%). This shows that even with the advances in medical technology and diagnostic methods, the rate of clinical misdiagnosis remained rather high (Table 2).
According to the disease categorization, the tumor misdiagnosis rate was 60.99%. Among the 4140 cases, 182 involved malignant tumors. Of these, 71 were correctly diagnosed (39.01%), 81 were misdiagnosed (44.51%), and 30 were undiagnosed (16.48%). The five most common locations of undiagnosed or misdiagnosed cancers were the brain (95.00%), endocrine system (92.30%), urogenital system (75.00%), blood (56.36%) and respiratory tract (46.67%) (Table 3). The discordance rate between clinical and autopsy diagnosis is similar to other report.2
Among the 4140 autopsies, there were 76 cases of deep fungal infection and the total misdiagnosis rate was 86.10% (Table 4). The rate of occurrence increased from 1950 to 1999: 0.24% in 1950–9, 2.51% in 1960–9, 4.03% in 1970–9, 3.06% in 1980–9, and 2.94% in 1990–9. From 2000 to 2009, there were only 4 fungal infection cases, the occurrence rate being 0.4%.
From 1950 to 1999, the autopsy rate declined steadily with a sharp decrease between 1990 and 1999 (Figure 1A–B), with an average of 10 cases per year, and the fewest being 5 cases in one year. The reasons for this decline are likely to be: (1) The social and hospital environment. The first refers to societal and traditional beliefs and habits, and the second to the fear of lawsuits brought against the hospital by the family of the deceased in the event of a clinical misdiagnosis confirmed at autopsy . (2) Clinicians may avoid using advanced medical technology to confirm the clinical diagnosis. (3) Lack of emphasis by the hospital and lack of initiative by clinicians. The fact is that the 10–15% autopsy rate required by the Ministry of Health only exists on paper and not in reality. Hospitals have not encouraged the development of autopsy work. (4) The fee for conducting an autopsy and the subsequent arrangements must be borne by the family of the deceased, though the resources for the autopsy are provided by the hospital.
Of these, 2785 were male and 1355 female. The largest number of cases occurred between 1960 and 1969. The fewest were recorded between 1990 and 1999(Figure 1A). The Adult and the infant groups are the major part in all cases (Figure 1B).
Within the 60 years, based on the disease distribution, respiratory diseases ranked the highest (1349, 32.58%), followed by circulatory diseases (495, 11.96%) and brain diseases (424, 10.24%), and then diseases of the alimentary tract (413, 9.98%) and infections (369, 8.91%). These five categories accounted for 73.67% of the total cases (Figure 2A).
In terms of the diagnostic error rate, the 2000–9 group was the highest. Second was the 1970–9 group, and third was the 1990–9 group (Figure 2B). In terms of the diseases, the most misdiagnosed were in the circulatory system. Ranking second among the categories with misdiagnoses was tumor, followed in descending order by parasitic diseases, and diseases of the genitourinary system, alimentary tract, and respiratory tract (Figure 2C).
In terms of the diagnostic error rate, the 2000–9 group was the highest (73.82%). Second was the 1970–9 group at 43.19%, and third was the 1990–9 group at 41.18%. Even for the lowest group, the error rate was as high as 30.63%, in 1980–9 (Figure 2B). In terms of the diseases, the most misdiagnosed were in the circulatory system (error rate 76.97%). These mainly occurred in patients whose cause of death was unknown at the time of autopsy. The rate of tumor misdiagnosis was 60.99% (Figure 2C), higher than the national average (26.29%) and that overseas (44.00%) . In the 182 cancer cases, the misdiagnosis rate ranked highest for brain cancers at 95.00% (19/20), followed by endocrine cancers at 92.30% (12/13), urogenital cancers at 75.00% (12/16), and blood cancers at 56.36% (31/55). Breast cancer (only 1 case) and the other cancers group (10/10) were not included (Table 3).
Ranking third among the categories with misdiagnoses was brain diseases (231/424, 54.48%), followed in descending order by parasitic diseases (31/63, 60.99%), and diseases of the genitourinary system (43/95, 45.26%), alimentary tract (183/413, 44.31%), and respiratory tract (536/1349, 39.37%) (Figure 2C). The parasitic diseases mostly occurred between 1950 and 1959 (29 cases). There were only 2 cases each in 1980–9 and 2000–9, and none in 1990–9. This reflects improvements in hygiene and a decline in occurrence in Hangzhou (Figure 3A).
The different misdiagnosis rate for the different diseases between 1950 and 2009(Figure 3B).
An improvement of the situation has occurred in recent years. In the most recent decade (2000–9), a total of 978 autopsies were conducted, nine-fold more than the 102 between 1990 and 1999. We found that cases of adult sudden death increased within the most recent decade. There were 72 in 2000–9, while there were only 118 in total from 1950 to 1999. Since most of the autopsies resulted from dissension between the relatives and the medical staff along with an unclear diagnosis before death, the misdiagnosis rate had evidently increased that decade. The misdiagnosis rate of 73.82% in 2000–9, the highest of all the decades, with 92.95% for diseases of the circulatory system (224/241), 86.28% for the respiratory tract (283/328), and 69.94% for the brain (121/173). The misdiagnosis rate was the highest for diseases of all three systems (Figure 3B).
In other major medical centers in China, the misdiagnosis rate was also high –. For example, the lowest rate was 20.00% reported in Guangzhou from 1958 to 1992. The findings from our analysis are similar to those reported by other groups (Table 5). The diagnosis technology has developed rapidly in the last sixty years, while the discordance rate has remained high in our cases. One reason is that many patients died because of sudden death with a short duration of hospitalization. In these cases, clinicians had to give the immediate and the underlying causes of death for patients dying under their care, sometimes lack of careful medical examination. That is one of the reasons that the discordance rate is particularly high for certain diseases, such as cardiovascular diseases. Adding CT/MRI/PETCT imaging results may help to reduce the misdiagnosis rate. The second reason is many cases came from grade I or grade II hospitals, whose medical levels are not as good as that of grade III hospitals. The misdiagnosis rate in grade I or grade II hospitals was higher than that of grade III hospitals. Thirdly, many cases consisted the diagnosis disputes between the relatives and the clinicians. At present, there exist increasing trend for medical malpractice disputes. In the case of medical disputes in the death, as the best evaluation and the objective and fair solution to the problem, the autopsy is necessary for doctors and families.
In US, the results of 53 distinct autopsy series over a 40-year period (from 1966 to 2002) show statistically significant decreases over time for major diagnostic errors detected at autopsy . In China Hangzhou, during the 60 years from 1950 to 2009, the discordance rate between clinical and autopsy diagnosis remained high. Although the autopsy rate decreased from 1950 to 1999, both the autopsy number and the diagnostic error rate increased during the most recent decade (2000–2009). With sufficient emphasis and initiative by the patients' relatives and doctors, the field of autopsy can be improved. This will be beneficial both for patients and their families , and the advancement of practice and research in clinical medicine in China.
Conceived and designed the experiments: KZ. Performed the experiments: KZ YX ZM WZ JC LM MC QS SZ. Analyzed the data: KZ HF. Contributed reagents/materials/analysis tools: KZ. Wrote the paper: KZ.
- 1. Balić V, Rumboldt Z (1995) Public attitudes toward autopsy, organ donation, and dissection. JAMA 273:1907.
- 2. Burton EC, Troxclair DA, Newman WP 3rd (1998) Autopsy diagnoses of malignant neoplasms: how often are clinical diagnoses incorrect? JAMA 280:1245–1248.
- 3. Zhu KQ, Zhang SJ (2004) Analysis of autopsy cases in 50 years. Zhonghua Nei Ke Za Zhi 43: 128–130. Chinese.
- 4. Li GD (1989) Study of the morbidity structure by analyses of data from 6668 autopsied cases. Zhonghua Bing Li Xue Za Zhi. 18:250–253.
- 5. Zhu MH, Yu DH (2011) Fluctuations in the rate of autopsy in China. Chin Med J (Engl) 124: 3403–3407. Review.
- 6. Shojania KG, Burton EC, McDonald KM, Goldman L (2003) Changes in rates of autopsy-detected diagnostic errors over time: a systematic review. JAMA 289: 2849–2856. Review
- 7. Burton JL, Underwood J (2007) Clinical, educational, and epidemiological value of autopsy. Lancet 369:1471–1480.