Analysis of cerebrospinal fluid (CSF) samples is a part of routine pediatric autopsies and is recommended in cases of sudden death of teenagers and young adults so that the possibility of meningitis can be eliminated. Additionally, analysis of CSF samples could be useful in many intracranial hemorrhage cases. CSF samples can be collected by performing lumbar or cisternal punctures; the latter is performed by accessing directly below the occiput through the atlanto-occipital membrane into the basal cistern. A spinal puncture could help diagnose intracranial hemorrhage, if blood is present in a CSF sample. Additionally, examination of postmortem CSF samples could determine hemorrhage dating [1]. However, postmortem diagnostic procedures, such as cisternal punctures to collect CSF samples, can cause local vascular injuries and hemorrhage; further, false-positive results could be obtained [1]. The authors performed cisternal punctures during routine autopsies; further, to evaluate the usefulness of the former, the results were analyzed.

In total, 233 autopsies were performed in the study period, and cisternal punctures were performed in 50 cases. These cases were reviewed. Results of cisternal puncture, autopsy findings such as head and neck injury, and hemoglobin concentrations of the aspirated fluids were evaluated. Evaluation of hemoglobin concentrations of the aspirated fluids was conducted by using AVOXimeter 4000 (Accriva Diagnostics, San Diego, CA, USA). CSF samples were collected by performing cisternal punctures, which included insertion of a tapping needle through the skin below the external occipital protuberance into the cisterna magna. Conventional autopsies were subsequently performed.

The age ranged from 1 day after birth to 82 years (average age, 44 years). Thirty-five male and 15 female dead bodies were studied. Postmortem interval was approximately 49.7 hours (minimum, 11 hours; maximum, 144 hours) according to autopsy requests. Postmortem intervals in inaccurate cases were decided by calculating time from detection of body to autopsy. Volume of aspirated fluids was a minimum of 1-2 mL. The authors did not aspirate large volume of fluids to avoid traumatic aspiration. Bloody fluid was aspirated in 20 cases, and CSF samples of the other 30 cases were clear. Acute hemorrhage of the head and neck was not observed in cases with clear CSF samples (Fig. 1). The main causes of death in cases with clear CSF samples were drowning, ischemic heart disease including acute myocardial infarction, sudden infant and adult death syndrome, and trunk trauma (Table 1).

Fig. 1 Clear cerebrospinal fluid is aspirated during cisternal puncture. The case number is 49, and the cause of death is aortic dissection with cardiac tamponade. Click for larger image

Table 1 Clear cerebrospinal fluid cases Click for larger image

Twenty cases in which bloody fluid was aspirated were divided into two groups. One group included cases with blood-tinged CSF samples (Fig. 2A, group A), and the other included cases of blood aspiration (Fig. 2B, group B). As determined by AVOXimeter 4000, hemoglobin concentrations of the CSF samples in group A and the aspirated fluids in group B were <4 mg/dL and >8.2 mg/dL, respectively. Groups A and B comprised eight and 12 cases, respectively. No injury in the head and neck was observed in six cases of group A; however, head injuries were observed in the remaining two cases. Main injuries were focal subarachnoid hemorrhage (SAH) and anterior cranial fossa fractures due to falling down (Table 2). No injury in the head and neck was observed in two cases of group B, and these results indicated traumatic tapping. The remaining ten cases of group B had head and neck injuries including non-traumatic basal SAH. There were three non-traumatic SAH cases, three head injury cases, and four cervical injury cases. Main head injuries were diffuse subdural hemorrhage (SDH) and middle/posterior cranial fossa fractures. Main cervical injuries were cervical dislocations and fractures (Table 3).

Fig. 2 (A) Blood-tinged cerebrospinal fluid is aspirated during cisternal puncture. The case number is 17, and the cause of death is drowning. (B) Bloody fluid is aspirated during cisternal puncture. The case number is 34, and the cause of death is non-traumatic subarachnoid hemorrhage. Click for larger image

Table 2 Group A cases (blood-tinged CSF group) Click for larger image

Table 3 Group B cases (blood aspiration group) Click for larger image

Examination of CSF samples is important during autopsies of children and adults. To collect CSF samples, both cisternal and lumbar punctures can be performed. Especially, examination of CSF sample in a pediatric autopsy is useful to determine the cause of death such as meningitis and sudden infant death syndrome [2]. CSF samples obtained during postmortem examination can be further tested; for example, bacterial culture tests and cell counts can be conducted [3]. Additionally, if bloody aspiration is observed during a spinal puncture, intracranial hemorrhage could be diagnosed. However, postmortem diagnostic procedures can cause local vascular injuries and hemorrhage; moreover, false-positive results might be obtained. The authors performed cisternal punctures during routine autopsies; further, to evaluate the usefulness of the former, the results were analyzed.

Out of 50 autopsy cases, clear CSF samples were aspirated in 30. Bloody aspiration was observed in 20 cases, and based on hemoglobin concentrations of the aspirated fluids, eight cases with blood-tinged CSF samples (group A) and 12 cases of blood aspiration (group B) were differentiated. Group A included cases of traumatic tapping; however, two out of eight cases had local intracranial hemorrhage. In contrast, group B included cases of true blood aspiration; however, two out of 12 cases had no hemorrhage in the head or neck during the autopsy. Results for group B were considered as positive; hence, sensitivity and positive predictive value of a cisternal puncture was 83.3%, and specificity and negative predictive value was 94.7%.

Head injuries identified in group A included focal SAH, vertex epidural hemorrhage (EDH) (distant from the skull base), and calvarial and anterior cranial fossa fractures. All of these injuries were caused by falling down. However, ten cases of hemorrhage in group B included three cases of basal SAH, three cases of head trauma, and four cases of neck trauma. All the three basal SAH cases were non-traumatic. Head injuries observed in group B were diffuse SDH in the skull base, and skull fractures were observed in the middle and posterior cranial fossa. Injuries identified in the four cases of neck trauma were primarily cervical fractures and dislocations.

This study showed that cisternal punctures performed for CSF sample collection could enable identification of head or neck injuries such as intracranial hemorrhage; these procedures showed high sensitivity and specificity and yielded positive and negative predictive values. Especially, in cases which clear CSF was aspirated, there was no hemorrhagic evidence in head or neck injuries during autopsies. This is showed a highly reliable negative predictive value. Furthermore, evaluation of hemoglobin concentrations of the aspirated fluids could decrease false positive results. If blood-tinged aspiration (group A) is observed, it could be interpreted that vascular injuries were sustained during the cisternal puncture per se. Further, forensic pathologists should be aware of the possibility of traumatic tapping in group B (blood aspiration group). Therefore it is recommended to carry out an additional lumbar puncture, if the bloody fluid is aspirated during a cisternal puncture. Out of the head trauma cases with intracranial hemorrhage, EDH or SDH (where the injury is far from the skull base) could represent negative cisternal puncture results; however, diffuse SDH (near skull base) could represent positive cisternal puncture results. With respect to the skull base fractures, negative result was observed for anterior cranial fossa fracture; however, positive results were observed for the middle and posterior cranial fossa fractures. Additionally, positive results were observed for all the cases of non-traumatic basal SAH; therefore, cisternal puncture is likely to be useful in cases of assumed natural death. Cervical injuries including fractures or dislocations of the cervical vertebra could be suspected with blood aspiration during the cisternal puncture. These results show that cisternal puncture is a useful procedure during postmortem examinations that can help to identify cervical injuries and intracranial hemorrhages including basal SAH.