Prevalence of neurogenic pulmonary edema among patients who died from head injury – a retrospective chart review

Neurogenic pulmonary edema is a less recognized complication Background: of head injuries and is seldom considered in management protocols in most centers. We therefore conducted a study to determine the prevalence of Neurogenic Pulmonary Edema among patients who died from head injury at Mulago National Referral Hospital, Uganda. An observational study conducted from the 1  June to 31  August Methods: 2013, with ethical approval. We consecutively enrolled patients who died of head injuries in the hospital. Demographic data, duration of illness, severity of illness, and patient management instituted were gathered from patient medical files. Autopsy results of the brain, lungs, heart, liver and kidneys performed within 24 hours after death were reviewed. Twenty-six patients who died from head injury were enrolled in this Results: study over the study period. Only one patient had preexisting hypertension and obesity while the rest had no known premorbid medical history. Intracranial abnormalities including raised intracranial pressure had been detected by CT scan in 18/26 of the patients while others had a normal scan (1/26) or did not have a scan done (7/26). Features of pulmonary edema were detected in 76.9% (20/26) of the bodies on gross lung examination. Pulmonary edema was bilateral in 85% of the bodies. Congestion was also noted in the liver, heart and kidneys. Pulmonary edema is highly prevalent in head injury patients and Conclusions: needs to be critically recognized early in the formulation of a management plan, as it contributes to morbidity and secondary brain injury through respiratory embarrassment.


Introduction
Neurogenic pulmonary edema (NPE) is an age-old clinical syndrome characterized by the acute onset of pulmonary edema following a significant central nervous system (CNS) insult 1 .NPE can occur after virtually any form of injury of the CNS, and is a potential early contributor to pulmonary dysfunction in patients with head injuries 2 .A myriad of CNS events, including spinal cord injury, subarachnoid hemorrhage, traumatic brain injury (TBI), intracranial hemorrhage, status epilepticus, meningitis, and subdural hemorrhage, have been associated with this syndrome [3][4][5][6] .
The exact mechanism is unknown as it is believed to be multi-pronged and data is sparse.The connection between CNS injury and hemodynamic dysfunction was first described by Harvey Williams Cushing in 1903 1 .This was followed by reports of cases in which acute pulmonary edema developed after CNS insult resultant from status epilepticus 1,7 and isolated head bullet wounds 3 .The etiology is thought to be a surge of catecholamines following the head injury that results in cardiopulmonary dysfunction 8,9 .Several cardiovascular events are likely to result from this stimulation resulting in capillary leak and eventual pulmonary edema 1 .Bahloul et al. identified features of left ventricular myocardial dysfunction in an echocardiographic study performed on previously healthy traumatic brain injury patients presenting with neurogenic pulmonary edema 5 .
TBI is one of the world's major causes of mortality and morbidity, particularly in the developing world, where motor vehicle and cyclist crashes are largely due to human error and poor roads [10][11][12][13] .With increasing industrialization of the developing world, there is a proportional increase in motorized trauma.Uganda is currently ranked second in the world in road traffic accidents after Ethiopia 14 .These in a country with a limited health budget, mean resources are stretched to the limit by the burden of care required by trauma patients.TBI forms the majority of trauma cases occurring, either in isolation or as part of multiple trauma 14,15 .
The neurosurgery Unit of Mulago NRH admits at least ten critically ill patients per day, while the intensive care unit (ICU) can only admit a maximum of four patients.The national critical care bed capacity is significantly limited for the enormous population in need 16 .An in-hospital mortality of 25% following severe TBI was reported in a retrospective study in Mulago NRH 13 while Kwizera et al. found an ICU mortality rate of 45.3% from head injury nationally.Mortality was commonly due to hemorrhagic shock, respiratory failure and sepsis 1 .
The majority of survivors are left with gross disabilities while out-of-hospital mortality post-discharge remains largely unknown 16 .
Awareness of NPE may lead to early intervention in braininjured patients.However, although NPE was identified over 100 years ago, it is still underappreciated in the clinical arena.Its sporadic and relatively unpredictable nature, and a lack of etiologic-specific diagnostic markers and treatment modalities, may in part be responsible for its poor recognition at the bedside 1 .
Worse still, there is very little data on the subject.Current data estimate the prevalence of NPE following severe brain insults, like trauma, stroke, status epilepticus etc., to be about 50% 17 , rising up to 92% in fatal subarachnoid hemorrhage 18 , being directly related to duration since insult 4 .Most data were, however, based on reviews of hospital records and case studies.
In this study, we determined the prevalence of neurogenic pulmonary edema through postmortem studies of head injured patients and sought to establish its relationship to the different patterns of organ dysfunction in head injured patients.This study also indirectly assessed quality of care of neurosurgery patients in the hospital.There are no published data on a similar study to the best of our knowledge.

Study design
This was an observational study carried out in Mulago National Referral and Teaching Hospital, Uganda, involving patients who died from head injury.We consecutively enrolled subjects into the study upon death over two months (1 st  Postmortem studies were performed by a pathologist (KS) within 24 hours of death.Full physical examination of the bodies was done, and notable abnormalities documented.Features of edema were sought during the examination of the heart, lungs, brain and kidneys.Edema was defined by tissue congestion with interstitial fluid on gross examination.A calibrated beam balance was used to weigh the brain, lungs, kidneys and heart.Organ weights were compared to standardized normal ranges based on the European guidelines 19

Data analysis
Basic descriptive statistics were used to analyze demographic data and other study variables.Data were analysed in STATA 10.0 Logistic regression analysis was used to determine the association between different variables with presence of NPE.P-values <0.05 were considered statistically significant (Table 3).Data are presented as modes and means unless otherwise indicated.1).Autopsy showed features consistent with pulmonary edema in 76.9% of patients, being bilateral in 85% of these.All the 15% in whom edema was unilateral had involvement of the right lung (Table 2).

During
Brain CT findings included intracranial edema and hemorrhage in 69.2%.14 (53.9%) of patients had concomitant brain contusions while only one did not reveal any radiological features of intracranial pathology.Only three (11.5%)patients had had chest x-rays performed which revealed features consistent with bilateral airspace disease (Table 1).All patients admitted to ICU received invasive artificial ventilation.Average daily total fluid input was 2 liters with an average overall measured fluid balance (total IV and oral fluid input minus outputs from drains (urine, GI) (Table 1) of +560 mL at time of death.No transfusion reactions were recorded.None of the patients required instrumentation of the chest wall.Five patients (30.8%) had received orthopedic surgical care, two (7.7%) had received surgical toilet for soft tissue injuries while an additional two patients had undergone exploratory laparotomy for blunt abdominal trauma (ruptured spleen and ruptured liver, respectively) (Table 1).
Features of pulmonary edema were noted in 76.9% of patients.85% of these showed evidence of pulmonary edema bilaterally.
Additional injuries noted at autopsy included liver and splenic contusion in one patient each, musculoskeletal injuries (degloving injuries (7.7%), long bone fractures (30.8%).There was no evidence of spinal trauma at autopsy (Table 2).Organs from one patient aged 9 years were not weighted against the European standard to which organ weights were referenced.

Discussion
The true incidence of NPE after acute head injury is difficult to estimate because much of the information comes from small autopsy series or isolated case reports.In this postmortem study, we found prevalence of neurogenic pulmonary edema of 76.9%.In an autopsy and in-patient database study, Rogers et al. found a prevalence of neurogenic pulmonary edema of   Pulmonary dysfunction after acute brain injury is a common but poorly understood phenomenon.In classic medical literature, the causes of pulmonary dysfunction in patients with head injury include pneumonia, aspiration, and pulmonary embolus, while NPE is seldom recognized.NPE is a form of pulmonary edema that develops rapidly after a cerebral injury 20 .It has been described in trauma patients as parenchymal edema, hemorrhage, and congestion without evidence of chest trauma in patients with isolated head injury 18,21 .
In Uganda, over 20,000 road traffic crashes occur annually claiming many lives through multiple trauma, head injury being one of the leading causes of death 14,15,22,23 .However, the incidence of NPE had not previously been established.In the study reported here, we included only patients with traumatic head injury who died after admission and treatment in the hospital.In addition, because of a shortage of resources, it was not possible to carry out several specialized tests on the patients before death.
Pulmonary edema was bilateral in most patients studied and was predominantly right sided in those with unilateral edema.This relates closely with the theory of a generalized pulmonary vascular disorder in head injured patients which is related to the severity of insult 24 .The study by Rogers et al., however, describes a predominantly right sided edema presentation 2 .This difference could be related to the difference in severity of illness and quality of initial management instituted.Furthermore, our study applied an European standard to organ weight measurement which was validated in a different population and only gross lung examination was performed for evidence of pulmonary edema which could have contributed to disparity in measurements and findings.
Studies on the other organs showed no significant weight increases.In addition, no significant structural abnormalities indicative of preexisting chronic conditions could be found on anatomical examination of these organs.This correlates well with the definition of neurogenic pulmonary edema in which there is no other significant organ injury in a patient with severe brain injury 1,24,25 .Edema was, however, seen in some dissections of the heart and kidneys that could be a result of multi-organ dysfunction as a result of critical illness.In addition, Bahloul et al. described myocardial dysfunction in neurogenic injury which could in part explain the pathophysiology of NPE 5 Severe TBI frequently presents with significant depression in level of consciousness compromising adaptive protective reflexes.Standard management for severe TBI patients includes ICU admission for intensive monitoring, interventions and supportive care such as invasive mechanical ventilation, among other therapies.Moreover permissive hypercapnia and tight control of blood carbondioxide tension forms an integral part of neurocritical care 26,27 .In our study however, only 26.9% of patients were managed in ICU.Consequently, only 26.9% of patients in this study had received mechanical ventilatory support despite 76.9% having potentially developed respiratory dysfunction due to pulmonary edema among other causes.This is partly explained by the shortage of ICU capacity nationally 16 and underscores the pressing need for critical care services in the care of TBI patients.
Brain CT scan is a minimum diagnostic tool.In our study, 26.9% of patients never had a brain CT scan performed.This in addition to limited surgical capacity leads to delay and missed opportunities.As a result, the mortality of head injury patients is very high, resulting perhaps in complications such as neurogenic pulmonary edema.Absence of complete antemortem investigation of the patients before death for pulmonary edema plus associated conditions as well as the small sample size makes it difficult to deduce statistically significant causal relationships in our study.However, strong inferences can be derived from association between severity of brain injury (as shown by need for ICU admission, type of injury) with NPE.

Conclusion
We conclude that NPE occurs frequently in head injury patients.The process of edema formation begins early in the clinical course and is isolated to the lung.NPE thus needs to be critically recognized early in the formulation of a management plan.The contribution of NPE to mortality and morbidity in these patients cannot be deduced from this study, as a causal association between pulmonary edema and mortality was not studied.

Ethical approval and consent
Approval of the study protocol was obtained from Mulago National Referral and Teaching Hospital Ethics Board.Written informed consent to participate and publish patient records was obtained from relatives.Denial of consent was criteria for omission from the study.All authors vouch for the originality, completeness, accuracy of the data presented and for the fidelity of this report to the study protocol as detailed herein and copied to the Mulago National Referral and Teaching Hospital Ethics Board.

Open Peer Review
Current Referee Status: Okello et al provide observational data on the incidence of pulmonary edema at autopsy following death from head injury.Consecutive data, of organ weight, with in 24 hours of death, obtained over two months, in 26 patients are provided.Pulmonary edema was documented in 76.8% of patients; weight increase was not found in organs other than the lungs.The mean age was 26 making co morbid cardiovascular factors unlikely.This incidence is similar to that of Simmons (1969) report in fatal military head injuries (pulmonary edema found in 17 of 20 cases).The authors conclusion that neurogenic pulmonary edema begins early and is isolated to the lung adds incidence data to a fatal consequence of head injury about which the pathophysiology and therefore the treatment remains uncertain.

Are sufficient details of methods and analysis provided to allow replication by others? Yes
If applicable, is the statistical analysis and its interpretation appropriate?Not applicable Are all the source data underlying the results available to ensure full reproducibility?No source data required

Are the conclusions drawn adequately supported by the results? Yes
No competing interests were disclosed.

Competing Interests:
I have read this submission.I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.

Are the conclusions drawn adequately supported by the results? Partly
No competing interests were disclosed.

Competing Interests:
I have read this submission.I believe that I have an appropriate level of expertise to state that I do not consider it to be of an acceptable scientific standard, for reasons outlined above.
The benefits of publishing with F1000Research: Your article is published within days, with no editorial bias You can publish traditional articles, null/negative results, case reports, data notes and more The peer review process is transparent and collaborative Your article is indexed in PubMed after passing peer review Dedicated customer support at every stage For pre-submission enquiries, contact research@f1000.com Department of Medicine and Neurobiology, Morehouse School of Medicine, Atlanta, GA, USA the study period, twenty-six patients were enrolled in this study.No patient had a premorbid diagnosis of renal, liver or lung disease.Only one patient had preexisting hypertension and obesity but with no documented cardiac disease.