15283-B-Blokker

86 Chapter 5 Diagnoses Within this study population 347 different post - mortem ICD diagnoses were encountered; of these, 230 (66.3%) were classified as major in at least one case (Table 3). From the MIA and CA reports of the 99 included cases, a total of 3097 post - mortem diagnoses were extracted (Appendix I); MIA identified 79.0% of these and CA 46.0% (Appendix J-1). 1372 (44.3%) of the diagnoses were initially classified as major; MIA detected 72.3% and CA 67.3% (Appendix J-2). After retrospective grouping, 85 different GMD remained. The post-mortem techniques together scored 283 GMD; 91.5% found by MIA and 79.2% by CA (Appendix J-3). In addition, there were five certain diagnoses in pre-mortem clinical evaluation that were not scored certain by either post-mortem method, but nevertheless classified as major diagnosis, because they were directly related to COD and therefore added to the list of GMD. Of the 288 GMD 124 (43.1%) were clinically unsuspected, 111 (89.5%) of the latter were found by MIA and 92 (74.2%) by CA; 79 were diagnosed by both MIA and CA, 32 only by MIA and 13 only by CA. Agreement between pre-mortem clinical evaluation, MIA and CA for all GMD, for GMD in lungs, heart, brain and vascular system, for pneumonia and myocardial infarction, and for all neoplastic diseases is illustrated in Venn diagrams (Figure 2). The performance of PMCT, PMMR, biopsies and CA separately in detecting GMD per category is shown in Appendix H. Inter - observer agreement (kappa) for radiological detection of GMD was 0.91 for PMCT and 0.80 for PMMR. Perceptual, Cognitive and Sampling Errors MIA made 16 perceptual errors: 12 on imaging and four on microscopic examination. There were seven cognitive errors: all on microscopy. Four diagnoses were missed due to sampling error. CA made 26 perceptual errors: nine on gross and 17 on microscopic examination. There were six cognitive errors. (Appendix K) CONCLUSIONS In this prospective study on a cohort of in-hospital deceased adult patients, MIA combining PMMR, PMCT, and image-guided biopsies, performed equally well as CA in identifying COD and answering specific clinical questions. MIA had a higher yield than CA for post-mortem diagnoses, major diagnoses and GMD, many of which were clinically unsuspected. The standard application of biopsies allowed making specific microscopic diagnoses with MIA, also of diseases not accompanied by obvious gross