Wet lungs, broken hearts and difficult therapies after subarachnoid hemorrhage

Pulmonary edema (PE) can occur after subarachnoid hemorrhage and can jeopardize arterial oxygenation, which is essential for a suffering brain. In some cases PE is evident in the emergency room, being the direct consequence of intracranial bleeding, which causes an immediate and overwhelming catecholamine discharge. In the following days, PE can occur because of cardiac failure, often related to initial cardiac damage, concurrent therapies with fluid overload and vasopressors, infections, or pre-existing co-morbidities. The causes of PE need to be identified for appropriate treatment.

An acute catecholamine discharge immediately following intracranial bleeding can have devastating extracranial eff ects: the heart, lungs, fl uids and electrolytes can be severely aff ected. Hoff and colleagues report on pulmonary edema (PE) following subarachnoid hemorrhage (SAH) [1]. Th ey found a high incidence of PE, usually several days after the initial bleeding, and an association of PE with lower intra vascular volumes (compared with cases without PE). PE is not uncommon after SAH, both as an early compli cation and as a late complication.
Neurogenic PE is an acute event directly linked with an intracranial disaster; as such, it is often diagnosed on admission in cases with severe SAH [2]. PE in the days following SAH has been reported in association with triple-H therapy, which includes hypervolemia and induced arterial hypertension [3]. In diff erent series, the PE incidence varied from 14 to 23% [4,5]. Increasing evidence is accumulating on the myocardial dysfunction that follows SAH, as well as other acute intracranial disasters. Takot subo syndrome -with typical left ventricular abnor malities (bulging out of the apex of the heart with preserved function of the base, which earned the syndrome its name -a kind of a pot used as octopus trap in Japan), ECG abnormalities and biomarker changesis a typical example of how much an injured brain can break the heart [6]. Neurogenic in origin, left ventricular failure causes congestion in the pulmonary vasculature and PE in Takotsubo syndrome.
What is concerning in the report of Hoff and colleagues is that PE, often labeled in the text as neurogenic PE, has been diagnosed days after the initial bleeding, without concurrent reduction of the cardiac index and in the absence of volume expansion [1]. Th e clinical implications of these fi ndings, as pointed out by the authors, would be that PE more than 4 days after SAH is not cardiogenic and would not be suitable for treatment with diuretics. It is therefore hard to reconcile the fi ndings of this paper with current knowledge.
Perhaps the methodology and terminology used in Hoff and colleagues' paper require further attention. PE has been identifi ed by clinical signs and bilateral pulmonary infi ltrates on the chest X-ray scan, without quantitative measures concerning the cardiac and respiratory function at the time of PE diagnosis. It is surprising that no data on oxygenation, such as a simple blood gas analysis, are reported. Th e hemodynamic status seems better docu mented, since the cardiac index and the circulating blood volume were estimated. Th ere are two weaknesses in these data, however, the fi rst of which is the reliance on a single technique without confi rmation by independent measures. Th e second, and more relevant, weakness concerns the data used in this specifi c analysis: the data do not represent the actual circulating blood volume and cardiac index at the time of PE. In fact, mean values calculated from day 1 to the day when PE developed have been entered in the analysis. It becomes almost impossible, therefore, to use these mean data for correctly understanding what caused PE. Interestingly, diuretics were used in 65% of patients with PE in the days before the PE diagnosis. Were they used for correcting an

Abstract
Pulmonary edema (PE) can occur after subarachnoid hemorrhage and can jeopardize arterial oxygenation, which is essential for a suff ering brain. In some cases PE is evident in the emergency room, being the direct consequence of intracranial bleeding, which causes an immediate and overwhelming catecholamine discharge. In the following days, PE can occur because of cardiac failure, often related to initial cardiac damage, concurrent therapies with fl uid overload and vasopressors, infections, or pre-existing co-morbidities. The causes of PE need to be identifi ed for appropriate treatment. excessive volume expansion, as seems likely? If that was the case, PE could have been cardiogenic, and/or somehow related to fl uid manage ment, without necessarily depending on a neurogenic cause.
Managing patients after severe SAH is challenging: on one hand, their brain requires normal (or even high, in the case of vasospasm) arterial pressure to warrant cerebral perfusion and prevent delayed ischemic defi cits; on the other hand, their heart may be damaged by an early sympathetic discharge, causing ischemic damage, and does not tolerate an increased workload. PE is a deleterious complication that may worsen systemic and cerebral oxygenation, and as such needs to be quickly recognized and treated [7]. Th e enthusiasm for triple-H therapy, which carries the risk of fl uid overload and an indiscriminate use of vasopressors, must therefore be tempered.
Perhaps the most important lesson to be drawn from this paper [1] is the need for accurate monitoring, both cardiovascular and respiratory, in all SAH cases. Some patients may simply require good clinical surveillance and careful fl uid balance. Th e more severe cases, on the contrary, should be cared for by a team capable of detecting early signs of heart failure, identifying the causes of PE and treating them, still preserving cerebral perfusion.