Delayed Posttraumatic Hemorrhage
Background and Purpose Delayed posttraumatic hemorrhage (DPTH) is one of the causes of intracerebral hemorrhage (ICH). Most of the published series were reported before the advent of CT. Clinical characteristics and CT findings are described.
Methods Ten cases of CT-proven DPTH from a series of 216 patients (age range, 15 to 50 years) with ICH were evaluated. Previous CT scans performed in the first 6 hours after head injury were normal in all patients. Cerebral angiography showed no evidence of vascular disorders in any patient.
Results DPTH was the fourth most common cause of ICH in our series. The symptom-free interval after head injury ranged from 1 to 15 days. Most hematomas were located deep in the hemisphere and were of small or medium size. Clinical course was good in the majority of cases; none of the patients died, and there was no recurrence.
Conclusions DPTH is a relatively frequent cause of ICH in young adults. There is no relationship with intensity of the previous head injury. The more frequent location deep in the hemisphere may be related to the younger age of our patients compared with those of other published series. The good outcome in our patients may be related to their youth and the absence of complications such as skull fracture, need for neurosurgery, or coagulation disorders.
At the end of the last century, Böllinger1 described four patients who died from an intracerebral hemorrhage (ICH) that occurred 2 weeks after they suffered head injury (“Traumatische Spät-Apoplexie”). Criteria proposed by this author for diagnosing delayed posttraumatic hemorrhage (DPTH), some of which are still valid today, were absence of previous vascular disease, history of head injury, presence of a symptom-free interval between the head injury and ICH, and subsequent stroke event.
Though not a rare cause of ICH, DPTH continues to be a little known condition. Most of the series published date from before the availability of CT, and there are very few cases of DPTH documented with CT. Present criteria have evolved since the advent of CT. Thus, Lipper et al2 diagnose DPTH if initial CT shows no lesions or if any lesion measures less than 1 cm and subsequent scans show high-density parenchymatous lesions. Fukamachi et al3 established that the initial CT scan must be performed within 6 hours after injury to allow DPTH to be distinguished from ICH of other causes.
The clinical and radiological characteristics and clinical course of a series of 10 patients with a diagnosis of DPTH are presented.
Subjects and Methods
Ten cases of DPTH were selected from a prospective series of 216 patients (age range, 15 to 50 years) with a diagnosis of ICH and admitted to the neurology department of our center. The following criteria were used for diagnosis: (1) a brain CT scan performed within 6 hours after head injury showing no abnormality, (2) no evidence of underlying vascular disease, (3) neurological impairment after an asymptomatic period, and (4) ICH appearing on a second CT scan.
All patients were treated in the neurosurgery emergency unit immediately after suffering the head injury. At least one brain CT scan was performed in all patients on admission to the neurology ward. A CT scan performed immediately after injury was also available for all patients. Cerebral angiography was carried out in all cases. Patient follow-up time varied between 14 and 86 months. The Oxford Handicap Scale4 (Table 1⇓) was used to assess functional ability at 6 months.
DPTH was diagnosed in 10 patients (4.5% of all patients with ICH) and was the fourth most common cause after arterial hypertension, vascular malformations, and alcoholism. Clinical features and neuroradiological findings are summarized in Table 2⇓. There were seven men and three women. Two patients were alcoholics, one of whom had high blood pressure. Two patients were smokers, and one was diabetic.
Head injury resulted from a traffic accident in three cases, one patient fell down stairs, and another fell from scaffolding. Two patients had a flex-extension mechanism, and one had a work accident. One patient suffered a bump on the head and another a blow with a ball.
Symptoms appearing immediately after the injury included concussion lasting less than 10 minutes in six cases, confusion in one, headache in one, and confusion and headache in one. Initial CT scans showed no sign of skull fracture, bleeding injuries, or areas of hematoma in any patient. The ICH occurred after a symptom-free interval ranging from 1 to 15 days after injury, with an average of 5±5.23 days.
The hematoma was located in capsule-basal ganglia in six cases, the frontal lobe in three, and pons in one. The hematoma volume ranged from 0.3 to 19.6 mL, with an average of 9.9±7.0 mL.
Clinical course was usually good. Severe neurological deficit (Oxford Handicap Scale score of 3) persists in only one patient. One patient suffered epilepsy. None of the patients died, and none suffered a second ICH during the follow-up period.
Etiologies of ICH in young adults, of which DPTH is one, vary greatly. Lipper et al2 diagnose DPTH if the initial CT scan shows no injury or if any injury measures less than 1 cm and subsequent scans show high-density parenchymatous lesions, excluding patients in whom the initial lesion has grown or developed in a site of previous surgery. The initial CT scan should be performed within the first 6 hours after the head injury and should show no signs of ICH. The nonavailability of CT or MRI beyond this period may lead the physician to diagnose a DPTH incorrectly, when in fact the ICH has another cause or the patient has a slowly developing traumatic hemorrhage.3
According to Fukamachi et al,3 posttraumatic ICH may be classified into four types: type I, hematomas visible on the initial CT scan; type II, small or medium hematomas that progressively increase in size; type III, hematomas developing in sites where the initial CT scan shows no hemorrhagic lesion; and type IV, hematomas forming in areas of contusion where the initial CT scan showed a “pepper and salt” image (areas of high density and isodensity). Unlike Lipper et al2 and Fukamachi et al,3 we consider that the initial CT scan should be normal to differentiate a slowly developing traumatic ICH (types II and IV in Fukamachi’s classification3 ) from a DPTH. Therefore, only type III would correspond to our definition of DPTH.
In our series, CT was performed in all patients within 6 hours of the head injury. A further CT scan was performed on admission to the neurology department. A neurosurgeon stated that there was no assessable neurological pathology in the first few hours after the head injury. CT performed after the head injury showed no assessable changes in any patient. Cerebral angiography was normal in all patients. A risk factor for cerebral hemorrhage was a long-term daily alcohol intake exceeding 80 g in two patients; no abnormalities in liver function or blood coagulation were detected in either. A further risk factor was arterial hypertension in one patient who also reported high alcohol consumption.
Depending on the series, the symptom-free interval between injury and appearance of signs and symptoms varies between 24 hours and several weeks, or even months.3 5 6 7 8 9 (Table 3⇓). In our series, the interval ranged from 1 to 15 days, with an average of 5±5.23 days.
With respect to the clinical features of DPTH, we detected no major departures from what has already been reported in the literature.3 5 6 7 The most important signs were appearance of a focal neurological deficit, slight impairment on the Glasgow Outcome Scale (drowsiness in one patient), and epileptic seizures in one patient.
Unlike the series of Baratham and Dennyson,6 in which the hematoma location was lobar in 19 of 21 patients, cerebellar in 2, and in the basal ganglia in 2, in our series of 10 cases the hematoma was located in the basal ganglia in 6, lobar in 3, and in the pons in 1 (Figure⇓). It should be mentioned that in the Baratham and Dennyson series, the age of patients ranged from 5 to 83 years, with the majority in their sixties, which is when the incidence of amyloid angiopathy as a cause of ICH should be considered. In our series, patient age ranged from 15 to 50 years, ie, young adults in whom amyloid angiopathy is unlikely to occur.
On the other hand, basal ganglia location is normally attributed to a possible hypertensive etiology. This was unlikely in our series because only one patient had this condition before injury, whereas both patients with hematoma in the basal ganglia in the Baratham series had high blood pressure.
Posttraumatic hemorrhage located in basal ganglia with discrete neurological signs and complete recovery has occasionally been described in relatively young patients. The underlying mechanism proposed for these cases would be rupture of the choroid anterior or lenticulostriate artery caused by violent acceleration-deceleration.10
On the basis of the cases described in the literature, prognosis is generally poor. Curiously, prognosis is similar in both the pre-CT and post-CT eras.11 In our series, prognosis was good in both the short and long terms. Six patients showed no residual neurological deficit, and one patient had sequelae consisting of epileptic seizures. Two patients had a minimum deficit. One had a mild deficit, and only one patient had a moderate-to-severe deficit resulting from the hemorrhage. There were no deaths. This may be due to our patients’ relative youth, the absence of other concurrent intracranial bleeding or skull fracture, and the facts that none of the patients required neurosurgery because of the head injury and that none had coagulation disorders.9 12
The pathogenesis of DPTH has not been definitively established. Bleeding is likely to occur in an area of the brain injured by the initial trauma. Brain scans obtained in patients who subsequently present with DPTH often show no abnormalities in the area where bleeding will develop. However, it is not known whether abnormalities exist that may predispose to DPTH and are beyond resolution by CT. Although little experience is available to date, studies using MRI may resolve this aspect in the future.11
According to the various hypotheses proposed, the initial injury would cause areas of necrosis in the cerebral parenchyma and walls of the small arteries1 due to vasospasm caused by the trauma13 and/or local metabolic disorders that would injure the vessel wall (release of free radicals, accumulation of lipid peroxidase).14 15 On the other hand, Kaufman et al16 postulate that the trauma would favor the release of thrombogenic substances from brain tissue, giving rise to localized intravascular coagulation and subsequent fibrinolysis. The result would be local vascular occlusion and infarction, with subsequent hemorrhage due to lysis of the intravascular clot. For their part, Evans and Scheinker17 and Gudeman et al18 propose the concept of dysautoregulation: loss of the capacity to self-regulate cerebral blood flow would favor the appearance of small perivascular hemorrhages, which on coalescing would form the hematoma. In laboratory experiments, vasodilatation has been observed in the venule-capillary union, with extravasation of blood and formation of microhemorrhages.19 It has also been postulated that the injury would cause venous congestion with subsequent venous exudation.8
Our patients presented a clear symptom-free interval and sudden onset of symptoms. No coagulation disorders or previous vascular abnormalities were detected. None of our patients presented any other type of ICH or required neurosurgery. We therefore consider that they closely fulfill the concept of DPTH as a clinical entity. DPTH is one of the causes of ICH in adult patients under 50 years of age. Most head injuries causing DPTH are of low intensity, and initial clinical signs are mild. Most DPTH are located deep in the hemispheres and are of small or medium size. Prognosis in the short and long term is good.
- Received February 28, 1995.
- Revision received May 29, 1995.
- Accepted June 12, 1995.
- Copyright © 1995 by American Heart Association
Böllinger O. Ueber traumatische Spät-Apoplexie: Ein Beitrag zur Lehre von der Hirnerschütterung. In: Hirschwald A, ed. Internationale Beiträge zur Wissenschaftlichen Medizin. Berlin, Germany: Festschrift, Rudolf Virchow; 1891;2:457-470.
Baratham G, Dennyson WG. Delayed traumatic intracerebral haemorrhage. J Neurol Neurosurg Psychiatry. 1972;35:698-706.
Nanassis K, Frowein RA, Karimi A, Thun F. Delayed post-traumatic intracerebral bleeding. Delayed post-traumatic apoplexy: “Spätapoplexie.” Neurosurg Rev. 1989;12(suppl 1):243-251.
Parodi CI, Cammarata S, Pizio N, Sacco G. Traumatic basal ganglia haemorrhage with slight clinical signs and complete recovery. J Neurol Neurosurg Psychiatry. 1992;55:72. Letter.
Tsubokawa T, Ueno Y, Kondoh T, Katayama Y, Moriyatsu N. Bleeding mechanism of delayed traumatic intracerebral hematomas [in Japanese]. Shinkei Gaisho. 1980;3:147-153.