Risk factors of deterioration in patients of head injury with non-operative management on first neurosurgical consultation

INTRODUCTION

Traumatic brain injury (TBI) is a leading cause of death and disability worldwide with India being one of the largest contributors to the numbers.^[1-3] In most of the tertiary care centers in India, neurosurgical opinion is sought with earliest possible resuscitation of a patient presenting with head trauma, to determine the further line of management. Early detection of non-neurological complications following TBI is very common which eventually determines the outcome.^[4,5] A computed tomography (CT) scan of the brain is the standard diagnostic tool in the determination of intracranial injury. Depending on the neurological examination and findings on CT scan of brain, the need of surgery is decided for acute TBI.^[6-8] The patients who do not require neurosurgical procedures initially are subjected to close observation under intensivists and neurosurgeons. Patients who do not improve with medical management or deteriorate clinically are routinely subjected to repeat CT scan and laboratory investigations to determine the cause of neurological deterioration.

We performed a retrospective analysis of risk factors, leading to neurological deterioration in conservatively managed TBI patients.

MATERIALS AND METHODS

This is a single-center retrospective analytical study. The study had been initiated after approval from the Institutional Ethics Committee. This study analyzed 275 patients with acute TBI and traumatic intracranial hematoma (tICH) on initial CT brain within 48 h of trauma, admitted under emergency trauma care intensive care unit (ICU) who did not require neurosurgical operation. The tICH included epidural hemorrhage, subdural hemorrhage (SDH), intraparenchymal hemorrhage, and subarachnoid hemorrhage (SAH). The study was conducted in King Edward Memorial Hospital, Mumbai, which is largest tertiary care center in West India.

Patients with following criteria were excluded: (1) Age <12 years, (2) on antiplatelets/anticoagulant medications, (3) head injury with normal CT study of brain, (4) patients with penetrating head injuries, and (5) admissions >48 h after the traumatic event. The criteria for non-operative management were primarily based on the clinical and radiographic findings on admission, including alert mental status, absent lateralizing signs, basal cistern effacement or obliteration, and midline shift <5 mm. Initial neurosurgical intervention was defined as an operation done immediately while the patient was in the emergency department’s ICU. Delayed neurosurgical intervention was defined as an operation done after the failure of non-operative management.

RESULTS

DISCUSSION

As per the World Health Organization (WHO), RTA is the sixth most common cause of death and will be the third leading cause of mortality and morbidity by 2030 in the developing countries.^[9,10] According to age groups, adolescents and adults showed high rate of traffic accidents while falls were the most common mechanism in the infants, preschoolers, and schoolchildren. The most common mode of injury was RTA (61.45%) in our study. Skull fracture was the most common injury in patients with mild and moderate TBI followed by extradural hematoma. In patients with severe head injury, skull fractures and SDH accounted for the highest proportions followed by sub-arachnoid hemorrhage and intraventricular hemorrhages.^[9] TBI is more prevalent in male population.^[10] However, there was no significant association of neurological deterioration with either age or gender of the patient in our study.

The management of TBI encompasses pre-hospital evaluation and care, in-hospital management, and continued follow up. The initial management of patients with TBI is identical to that of all trauma patients, focusing on the Advanced Trauma Life Support principles of the management of airway, breathing, and circulation. The primary goal of pre-hospital and in-hospital airway assessment and hemodynamic stabilization is to prevent hypoxia and hypotension which are known to be major causes of secondary brain insult. Hypoxia is defined as PaO₂<60 mmHg and hypotension as systolic blood pressure <90 mmHg. Endotracheal intubation in patients with TBI is indicated in patients with a compromised airway, SpO₂<90 despite supplemental oxygen, and a Glasgow Coma Scale (GCS) score of <9.^[7,11-13] Patients with TBI should be presumed to have spine injuries and immobilized during transport. A detailed history and examination of injuries and neurological assessment are followed in the emergency department.

Alcohol abuse has been known to be associated with increased morbidity and mortality among patients of TBI due to a blunting of sympathetic response.^[11,14] Around 28.36% of our patients had a history of alcohol consumption before the traumatic event. However, no significant association between the consumption of alcohol and neurological deterioration was seen in our study. The effect of alcohol on trauma outcome has been studied with mixed results.^[14-16] The mechanisms whereby alcohol intoxication reduces the mortality of head injury are still unknown. The diuretic effect of alcohol might reduce the intracranial pressure after trauma. In addition, brain atrophy in some chronic alcoholic cases may have more space for hematoma to prevent increased intracranial pressure.^[15,16] In addition, most of our patients had motorcycle-related injury, which might result in the lower injury severity score compared to other studies.^[15,16] Further study on the relationship between the level of blood alcohol and mortality is needed to elucidate the mechanism.

The brain, through autoregulatory mechanisms, maintains the cerebral blood flow over a wide range of blood pressures. This capacity is, however, deranged in approximately one third of the patient of acute TBI. PHI has been shown to be closely associated with neurological deterioration in patients of TBI worldwide with fivefold worsening of prognosis. PHI has been defined as worsening picture on serial CT brain due to the development of a new lesion or increase in volume of a previous lesion by more than 25%.^[17] Few authors also stated that incidence of PHI in head trauma patients was 42.3%.^[17] PHI has been found to be associated with deranged coagulation profile, older age, male gender, hyperglycemia, raised D-dimer, and increased time from injury to first CT scan.^[18] In this study, PHI was seen in 18.9% of the patients and among them 86.5% of patients showed neurological deterioration due to progression of lesion in serial scans without any association of other factors. There have been various prognostic models to predict the occurrence of PHI in patients of acute TBI based on admission variables. These include increasing age, prothrombin time more than 14 s, low level of initial consciousness, platelet count between 1 and 1.5 lakhs/mm₃, raised D-dimer, intra-axial bleeding, and midline shift on CT brain of 5 mm or more. These have been studied by Yuan et al. to be independent risk factors of progressive hemorrhagic injury.^[19] In our study, we have taken only platelet count and INR as a coagulation markers. However, we did not find a statistically significant association between low platelet count, deranged INR, and neurological deterioration. Majority of patients of our study had mildly deranged INR (1.5–2.0). Further, detailed study may analyze the effect of severe thrombocytopenia, high INR, and other coagulation parameters on head injury patients in our scenario. Moreover, it is noteworthy that prognostic models that predict the occurrence of PHI are not part of the initial assessment of patients of TBI in many trauma centers in India. Several factors including logistics and resource allocation limit the widespread use of these models. As the reported rate of PHI varies from 20 to 60%^[19] and since we report a high proportion of this group requires operative intervention, it is imperative for intensivists and neurosurgeons around the world to develop and validate prognostic models to determine the likelihood of PHI.

Furthermore, in a tertiary care center in an American state, SIRS was found to be a predictor of unfavorable outcome in patients with hemorrhagic brain injury presenting as sepsis or pneumonia in 50–75% of patients with isolated head trauma.^[20] Inflammatory mediators release after TBI is responsible for initiating and propagating early, delayed, and systemic effects of brain injury. Damage to the blood– brain barrier exposes the normally immune privileged brain to peripheral inflammatory mediators. Once systemic inflammation is triggered, it is subsequently followed by coagulopathy and complement deficit which may further exacerbate the bleeding associated with TBI.^[21] Secondary brain insults such as hypoxia and hypotension have been shown to aggravate the SIRS. Alternatively, the counter anti-inflammatory response of the body may result in immune suppression and make the patient prone to superadded infections such as pneumonia and urinary tract infections, especially in patients with known immune-compromised states.^[21,22] Musculoskeletal, chest, and abdominal injuries giving rise to significant systemic responses may also aggravate the neuroinflammatory responses to trauma by the aforementioned pathways.^[23] Trauma triggers a complex cascade of events that can be divided into hemodynamic, metabolic, neuroendocrine, and immune responses, leading to a multifocal pathophysiological process. However, inflammation is not in itself detrimental. Inflammation is necessary for the removal of challenges to the organism and subsequent restoration of homeostasis. However, hemorrhage and trauma can induce a dysregulated acute inflammatory response that affects several organ systems and sets in motion a vicious cycle of inflammation and damage driven by chemokines, cytokines, and products of damaged, dysfunctional, or stressed tissue. Activation of early phase inflammatory mediators and complement system gives rise to a generalized inflammatory state – SIRS.^[22] We found significant correlation between SIRS and neurological deterioration in our study.

Infective causes of SIRS, in our study, included pneumonia, urinary tract infection, central line-associated bloodstream infection, and wound infection. Pneumonia is a common non-neurological complication of TBI resulting from prolonged mechanical ventilation and pulmonary inflammation. It results in hypoxemia, fever, and hypotension which further aggravate secondary brain injury. Strict antisepsis, adequate pulmonary toilet, avoidance of urosepsis with timely change of urinary catheters, proper handling of central and peripheral venous catheters, and training of health-care officials for the prevention of cross infection are utmost essential for the avoidance of infective causes of SIRS in patients of TBI, which, as we report, is responsible for neurological deterioration in a significantly large proportion of our study population. Proper waste handling and administration of antimicrobials as per drug sensitivity reports will help overcome infective complications.

Serum sodium abnormalities have been found as a sequel of brain injury due to the pivotal role played by the brain in sodium balance. The use of diuretics, intravenous fluids, syndrome of inappropriate secretion of antidiuretic hormone (SIADH), and cerebral salt wasting syndrome (CSWS) have also been found to be associated with dyselectrolytemia. Hypernatremia (>150 mEq/L) may occur either through free water loss (lack of access to free water, diabetes insipidus, or mannitol) or sodium gain due to administration of hypertonic saline. As such, control of intracranial pressure remains an important factor in the management of patients with TBI. Mannitol is the pharmacological agent of choice for the management of increased intracranial pressure. However, concerns over intravascular volume depletion and renal failure have led to the emergence of hypertonic saline as a therapeutic option. Multiple studies have demonstrated an increased mortality in patients of TBI with hypernatremia. In our study, electrolyte imbalance was seen in 24.36% of patients of TBI. After adjusting the effect of SIRS, cerebral edema, age, sex, alcohol, platelet, and INR by applying logistic regression on neurological deterioration, we did not find significant correlation (P = 0.173). It is also worth noting that a higher proportion of patients with dyselectrolytemia had hyponatremia rather than hypernatremia. Patients with neurotrauma have hyponatremia resulting from SIADH and CSWS. Hyponatremia is detrimental in patients with TBI as it causes exacerbation of cerebral edema and intracranial pressure. On the contrary, hypernatremia in patients of TBI is said to occur most likely secondary to diabetes insipidus, and interestingly, post-administration of hypertonic saline. Mortality associated with hypernatremia is reported to be 40–60%.^[24] Hence, aggressive efforts need to be invested to maintain electrolyte balance to prevent neurological deterioration. However, in our study, we could not find significant correlation which may be due to early detection of dyselectrolytemia, serial monitoring, and its aggressive correction. We use 3% NaCl as a decongestant in raised intracranial pressure which is also beneficial in correcting existing hyponatremia. On the other hand, only few patients in our study had hypernatremia which allowed us to use 3% NaCl as a primary cerebral decongestant.

According to the results of our study, cerebral edema, both perilesional and generalized, has a positive correlation with neurological deterioration in univariate analysis but not in logistic regression test. Ten times higher mortality among patients of non-penetrating TBI at a Level 1 trauma center was documented with cerebral edema.^[25] As the severity of TBI increased from moderate to severe, the proportion of patients suffering from PHI, SIRS, and electrolyte imbalance increased in our study. Early identification and correction of risk factors associated with poor outcome and neurological deterioration in patients with TBI can lead to reduction in mortality among this group of patients.

About 6.18% of patients required decompressive craniotomy in our study. A large proportion of patients who deteriorated neurologically during their course of admission recovered with diligent monitoring, early correction of electrolyte imbalance, treatment of simultaneous infections, avoidance of hypothermia, and treatment of coagulopathy and acidosis. However, 19.27% of patients died whose etiology could not be depicted due to hemodynamic instability. Our study tries to highlight the management difficulties of TBI in the developing countries.

There are some limitations of our study. We could not assess the true effect of hypoxia and hypotension on SIRS in our study due to the confounding impact of initial resuscitation and hemodynamic stabilization. Non-infective causes of SIRS like neurogenic pulmonary edema were also not assessed in our study. Other coagulation factors like thrombin time have not been studied in our study to establish a definite relation with PHI.

Our center is a classical representative of trauma care ICU of developing world which caters services to the majority of population. Intracranial pressure monitoring, jugular venous oximetry, cerebral microdialysis, brain tissue oxygen tension monitoring, pressure reactivity index, and thermal diffusion flowmetry for the measurement of parameters such as cerebral blood flow, metabolism, and oxygen delivery are yet to be introduced in our center. These parameters are being studied in very advanced trauma ICUs of developed world. However, these techniques are yet to withstand the financial and logistic realities of the world. Authentic evidence is required before universal introduction of these parameters into health-care systems worldwide in the management of patients of TBI. Our findings may be extrapolated to the realities of vast majority of centers in India and the developing world.

CONCLUSION

We report severe TBI, PHI, and SIRS as strong independent predictors of neurological deterioration in patients of acute TBI. It is, therefore, imperative to keenly observe patients for signs of electrolyte imbalance and SIRS and correct the same in a timely manner to avoid a point of no reversal in these patients.