|Home | About | Journals | Submit | Contact Us | Français|
Malaria is a prime public health threat in developing countries like India. There is an unmet need of a simplified methodology for the purpose of triage and provision of intensive care to the severely infected patients in these areas.
We did a prospective study in patients (n=60) admitted with severe malaria in a single tertiary care center in the state of Haryana, India. We assessed the role of coma acidosis malaria (CAM) score in these patients when predicting mortality and morbidity events. Stepwise logistic regression analysis was applied to identify patients requiring intensive care based on the CAM score, and the prediction value of the scoring system was assessed among these patients.
Cerebral malaria (measured using the Glasgow coma scale) and acidosis (base deficit) were the major determinants of the CAM score. Serum bicarbonate levels and respiratory rates were assessed as the proxy markers of the base deficit as it is not always available. Morbidity increased steadily as the CAM scores increased. Sensitivity and negative predictive value of 100% depicted that the scoring system was able to identify patients who needed intensive care and accurately exclude the patients who could be conservatively managed in the ward. Positive predictive values of 73.9%, 68%, and 80.9% indicated that CAM, bicarbonate-based CAM (BCAM), and respiratory rate-based CAM (RCAM) scores, respectively, could precisely predict the morbidity and mortality events among patients with CAM scores ≥2.
CAM scores have precise predictive values in assessing patients with severe malaria. The scoring system helps in accurate and systemic triage of patients, irrespective of species, and directs the treating physicians toward vigilant treatment and emergency care.
Malarya Hindistan gibi gelişen ülkelerde önemli bir halk sağlığı sorunudur. Bu bölgelerde ağır bir şekilde enfekte olmuş hastalara triaj ve yoğun bakım sağlanması için basitleştirilmiş bir yönteme ihtiyaç vardır.
Hindistan’ın Haryana eyaletinde tek bir üçüncü derece sağlık bakımı merkezine başvuran ağır malaryalı hastalarda (n=60) prospektif bir çalışma yaptık. Bu hastalarda mortalite ve morbiditeyi tahmin etmede Koma Asidoz Malarya (CAM) skorunun rolü değerlendirildi. CAM skoruna dayanarak yoğun bakım gereken hastaları belirlemek için aşamalı lojistik regresyon analizi uygulandı ve bu hastalarda skorlama sisteminin tahmin değeri değerlendirildi.
Serebral malarya (Glaskow Koma Ölçeği ile ölçüldüğü gibi) ve asidoz (baz açığı) CAM skorunun başlıca belirleyici faktörleriydi. Serum bikarbonat ve solunum sayısı, baz açığı her zaman mevcut olmadığından dolayı, baz açığının proxy belirteçleri olarak değerlendirildiler. Morbidite, CAM skorları arttıkça sabit olarak yükseldi. Sensitivite ve %100 negatif prediktif değer skorlama sisteminin yoğun bakıma ihtiyaç duyan hastaları belirleyebildiğini ve klinikte geleneksel yollarla yönetilebilen hastaları doğru bir şekilde ekarte edebildiğini ortaya koydu. %73,9, %68 ve %80,9 pozitif prediktif değerler, sırasıyla CAM, BCAM ve RCAM skorlarının CAM skorları ≥2 olan hastalarda morbidite ve mortaliteyi tam olarak tahmin edebildiğini gösterdi.
CAM skorlarının ağır malarya hastalarının değerlendirilmesinde kesin prediktif değeri vardır. Skorlama sistemi cinsiyete bakmaksızın doğru ve sistemik triaj sağlar ve tedavi eden doktorları tedbirli bir tedaviye ve acil bakıma yönlendirir.
India is primarily characterized by unstable malaria transmission characteristics; it is seasonal with increased intensity related to the monsoons. Malaria mainly affects the poor population in remote rural areas, which is due to inadequate control measures and healthcare. Plasmodium falciparum is responsible for a majority of malaria-related morbidity and mortality .
The World Health Organization (WHO) established the criteria to diagnose severe malaria in 1990. The criteria were revised in 2010 to include clinical manifestations and laboratory values that portend a poor prognosis based on various clinical studies . Cerebral malaria, pulmonary edema, acute renal failure, severe anemia, and bleeding are some of the major complications of severe malaria. Immediate treatment must be instituted to the patients who have signs and symptoms indicative of such complications. Female sex, age greater than 65 years, pregnancy, non-immune status, coexisting medical conditions, no antimalarial prophylaxis, delay in treatment, and severity of the illness at admission (e.g., coma, acute renal failure, shock, pulmonary edema, and coagulation disorders) are the risk factors indicated in various clinical studies .
There is an unmet need of a simplified methodology for the purpose of triage and provision of intensive care to severely infected patients in these areas. Although WHO has framed the criteria to define severe malaria, it is too broad a tool to be used in the triage of these patients. As the predictive value of each of the factor included in the criteria is variable, it is difficult to gauge the mortality in such patients [4, 5].
Tropics have adapted the criteria for severe malaria defined by the WHO [6, 7]. However, it is difficult to predict deteriorating prognosis solely based on this criterion. Moreover, there was no differentiation between the predictability of prognosis between patients with single and multi-organ failure. Studies are needed for better triage of patients with malaria, and the development of better scoring systems for prognostication is imperative. Therefore, the present study is being undertaken for predicting the outcome of severe malaria in adults by using coma acidosis malaria (CAM) score.
Between May 2014 and October 2015, we did a prospective trial in 60 patients with complicated malaria in Pt. B. D. Sharma PGIMS, Rohtak, which is a tertiary healthcare center in the state of Haryana, India. Patients aged >16 years who fulfilled the modified WHO criteria for severe malaria were included in our study . The criteria of diagnosis were extended to include patients infected with plasmodium vivax malaria, and CAM scores were applied in these patients to predict the prognosis. We excluded patients with diseases other than severe malaria causing metabolic acidosis, decreased hematocrit, jaundice, renal failure, hypoglycemia, and respiratory distress.
Written informed consent was obtained from the patient or the guardian/relative, and the study was approved by the institution’s ethical committee.
The patients who fulfilled the modified WHO criteria for the diagnosis of severe malaria were evaluated. The confirmation of diagnosis of malaria was done with HRP2/pLDH-antigen-based rapid test (Bio Standard Diagnostics Pvt. Ltd.) and peripheral blood smear examination, and an independent sequential record of all the patients assessed was maintained . Coma acidosis malaria (CAM) score was applied to these patients. The arterial blood gas analysis of the patients was done in an automated blood gas analyzer, and the serum bicarbonate level was assessed. The base deficit of the serum bicarbonate values was obtained by subtracting the patients’ values from the normal value (24 mmol/L). The CAM score (0–4) was calculated as the base deficit score (0–2) plus the Glasgow coma scale (GCS; 0–2).
The assessment of usefulness of a modified CAM score using plasma bicarbonate level as the marker of acidosis, namely, bicarbonate-based CAM (BCAM), was also done. To obtain the acidosis component of the CAM score, we used cutoff values ≥24 mmol/L (score 0) for normal, <24 mmol/L to 15 mmol/L (score 1) for deranged, and <15 mmol/L (score 2) for very deranged. BCAM score (0–4) was calculated as the bicarbonate score (0–2) plus the GCS (0–2).
The respiratory rate-based CAM (RCAM) score was calculated by using the respiratory rate as a surrogate marker for acidosis. The cutoffs values were <20 breaths/min (score 0) for normal, 20–39 breaths/min (score 1) for deranged, and ≥40 breaths/min (score 2) for very deranged. RCAM score (0–4) was calculated as the respiratory score (0–2) plus the GCS score (0–2).
Patients who had a CAM score <2 were conservatively managed, wherein their renal profile was vigilantly monitored, and those who had a score ≥2 with hypoxia were intensively managed. All the patients who were admitted in the intensive care unit had hypoxia. By using hypoxia as a proxy indicator for the intensive care unit (ICU) admission, the sensitivity, specificity, and predictive values were calculated.
Data collected in the study were statistically analyzed using Statistical Package for the Social Sciences for Windows version 20 (IBM Corp.; Armonk, NY, USA). Here p<0.05 was considered significant. Continuous data were represented as mean value±standard deviation. Chi-square test and Fisher’s exact test were used to analyze the quantitative data. Bivariate correlations among different variables were studied using Spearman’s test. Bivariate logistic regression analysis was done to assess the significance of CAM scores in predicting the outcome of patients having severe malaria.
Among the total of 60 subjects included in the study, the age of patients ranged from 16 to 82 years with an average age of 34.3 years. Most patients were in the age group of 21–40 years. The patient population was predominated by males (61.7%). One female was pregnant at the time of admission. In the etiological distribution, 33 (55%) patients were infected with plasmodium falciparum species and 27 (45%) patients were infected with plasmodium vivax species (Table 1).
Among the 60 patients, 11 (18.3%) patients had hemoglobin of less than 5 g/dL with mean hemoglobin of the study subjects being 9.56±3.21 g/dL. Leukocyte count >12000/µL was present in 10 (16.7%) patients and 20 (33.3%) patients had thrombocytopenia with a count of less than 50×103/µL. The mean serum creatinine level for the study subjects was 3.84±3.38 mg/dL with more than 3 mg/dL being present in 35 (58.3%) patients. Urine RBC level of more than 2/HPF (criteria for hematuria) was present in 42 (70%) patients. A parasitic index of >2% was found among 23 (38.3%) patients. Deranged liver parameters indicated by increased liver enzymes, more than 3 times the reference value, were seen in 21 (35%) patients.
Serum bicarbonate values revealed that 14 (23.3%) patients had a value of less than 15 mmol/L and 46 (76.7%) patients had a value in between 15 and 24 mmol/L. The base deficit was calculated; the values were less than 2 in 4 (6.7%) patients, in between 2 and <10 in 46 (76.7%) patients, and ≥10 in 10 (16.6%) patients. Oxygen saturation of <93% was present in 17 (28.3%) patients, which indicated the presence of hypoxemia. Patchy infiltrates suggestive of pulmonary edema were present in 4 (6.7%) patients (Table 2).
There was a statistically significant association and correlation between parasite index ≥2%, serum creatinine >3 mg/dL, and CAM scores (p<0.05). Leukocytosis (leukocyte count >12000/µL) exhibited statistically significant association and correlation between CAM, BCAM, and RCAM scores (Table 3). Epidemiological factors such as age, gender, and body mass index did not significantly influence the CAM scores in predicting the outcome events of severe malaria. Clinical findings including hepatomegaly and splenomegaly had no effect on predicting morbidity and mortality events. Laboratory findings including anemia (hemoglobin <5 g/dL), thrombocytopenia (platelet count <50,000/µL), and increased liver enzymes (greater than thrice the reference value) had no independent relationship with CAM scores in determining the incidence of adverse events in patients with severe malaria.
The determination of species infecting the study population did not influence the outcome events. Although the WHO criteria has been framed to diagnose patients infected with falciparum species, in our study, there was a similar rate of complications that occurred in patients infected with plasmodium vivax species.
The outcome events of the present study were divided into two categories: conservatively managed patients in the ward by monitoring renal functions and patients who needed intensive care representing higher morbidity incidence and who died secondary to complications of severe malaria. Among the study population, there were 31 (51.7%) patients who were given ICU care, and one of them expired secondary to cerebral malaria and renal and respiratory failures. Among the patients admitted in the general ward, 2 patients succumbed to death. One patient had cerebral malaria and respiratory failure and the second had combined cerebral malaria and respiratory and renal failures. Although these patients required immediate ICU care, we could not provide it due to late presentation and unavailability of beds at the time of admission.
All the patients who were admitted to the ICU and the 3 patients that succumbed to death had CAM scores ≥2. Among 8 (13.3%) patients who had a CAM score =2 and were conservatively managed in the ward and recovered, 5 (8.3%) patients had renal failure and were given hemodialysis. The total number of patients who were conservatively managed in the ward and recovered after hemodialysis was 10 (16.6%) patients. The statistical correlation of the outcome events with that of the CAM scores was significant (Table 4).
Hypoxia with oxygen saturation <93% was used in the study as the proxy criteria for ICU admission. The sensitivity of CAM scores with respect to ICU admission was 100%, revealing that a score ≥2 could accurately identify patients requiring ICU care. The specificity of the CAM score in identifying true negatives was 86.04%, implying that out of the 43 patients who did not have hypoxia at presentation, 37 patients had a CAM score <2. The remaining 6 patients had a CAM score ≥2, but they did not require ICU admission because hypoxia was absent. The positive predictive value (PPV) of the CAM score, which indicates the occurrence of morbidity and mortality in patients, with value ≥2 was 73.9%. The negative predictive value of the scoring was 100%, which indicated that patients who had CAM scores <2 never required ICU admission.
Similarly, the sensitivity and negative predictive value of BCAM and RCAM scores were 100%. The specificity of BCAM and RCAM scores was 81.39% and 90.69%, respectively. The PPVs in assessing the occurrence of morbidity and mortality events in patients with CAM scores ≥2 admitted to ICU care were 68% and 80.95% for BCAM and RCAM scores, respectively.
Among tropical countries, complicated malaria has significant socio-economic impact. Reduction in morbidity and mortality events could be achieved by early diagnoses and timely treatments. The diverse manifestations of the disease depend on various host, parasitic, and environmental factors. A proper indicator for the prediction of severity should, therefore, be designated to allocate more resources to severely ill patients. Sophisticated and time-consuming scoring systems currently available are not applicable in resource-limited settings. The WHO criteria for severe malaria have been adapted for tropical regions. However, to predict the outcome of severe malaria and for proper triage of patients toward resourceful settings, there is an emerging requirement of a simple scoring system.
Hanson et al. , in 2009, proposed the idea of CAM scores, a 5 point scoring system, using GCS and base deficit marking for cerebral malaria and metabolic acidosis, respectively. By using the data from the South East Asian Quinine Artesunate Malaria Trial (SEAQUAMAT) study (the largest ever clinical trial involving adults with severe malaria), a logistic regression model was built to identify independent predictors of mortality among adults with severe malaria . The data thus obtained was validated in 2 series from Vietnam and Bangladesh [10, 11]. The usefulness of a modified CAM score using plasma bicarbonate level as a marker of acidosis (BCAM) was assessed for patients in SEAQUAMAT as the base deficit was not always available. As the respiratory rate in the 3 datasets showed a weak correlation with the plasma base deficit (p<0.001), the RCAM score was calculated for patients in the SEAQUAMAT dataset (n=867). A high PPV and practicality of the scoring system made an intense impact in identifying patients having severe malaria and treating them immediately.
The objective behind planning this study was to assess the prediction value of CAM scores in the Indian population as the scoring system is simple and easily applicable in emergency departments. All the previous studies were undertaken in patients infected with falciparum malaria. There has been a rising trend in vivax malaria presenting with severe complications. In this study, efforts have been made to assess the use of CAM scores even in vivax-infected patients and its value in prediction of the outcome.
In the etiological distribution, falciparum infection (55%) exceeded vivax infection (45%). According to the study by Saravu et al. , plasmodium falciparum was found to be more likely to cause complicated malaria than plasmodium vivax. Hyperparasitemia was found to be the determinant of pathophysiology and higher virulence in plasmodium falciparum. Even though many other studies support this notion, in the present study, on comparing both the parasites with CAM scores, the difference in the outcome events was not statistically significant. According to study conducted by Kochar et al.  in 539 adult patients diagnosed with severe malaria, vivax malaria had a similar impact on the mortality and morbidity events as that of falciparum malaria. The property of vivax infection to remain latent in the livers of infected patients retains it for prolonged periods even after its transmission is interrupted. This also explains the recurrence of the vivax infection in such patients . In the present study, the determination of parasite species (falciparum/vivax) did not influence the prediction of outcome events of severe malaria in patients. This is in accordance with the previous studies and recent literature [15, 16].
Leukocytosis, as well as leukopenia, has been reported in falciparum malaria. Abdalla et al.  showed that 29% of their patients with falciparum malaria and severe anemia had leukocytosis. Mild leukopenia has been described in uncomplicated malaria, but a neutrophil leukocytosis is an important abnormality in patients with severe falciparum malaria and is associated with a bad prognosis . In the present study, there was a significant association as well as correlation between increased leukocyte count and CAM scores. Leukocytosis is an independent entity in determining the severity of complicated malaria, and it has a significant influence in the prediction of severe malaria in adult patients.
As the transmission of malaria in the studied area is low, parasitic index >2% was used in the study as the definition of hyperparasitemia . In our study, 23 (38.3%) patients had parasitic index >2%. The statistical analysis using the chi-square test revealed that the association of parasitic index had a highly significant association with the scoring system in predicting the outcome events in these patients. The results of the present study have been supported by a previous study conducted by Rajkumar et al. , where the parasitic index effectively correlated with the severity and course of disease, and complications and mortality were seen more often in patients with a higher degree of parasitemia.
In the study by Godse RR involving the hematological and biochemical evaluations in malaria patients with a clinical correlation, there was a concurrent increase in the serum creatinine level in patients with severe malaria, which was attributed to impaired glomerular filtration of urea and creatinine . There are other studies in the literature that have also reported high levels of creatinine in children and adults with severe malarial infections [22, 23]. In our study, serum creatinine >3 mg/dL (WHO criteria for the diagnosis of severe malaria) had a statistically significant association and correlation with CAM scores used in the prediction of outcome events in patients with severe malaria. In addition, blood urea levels also had highly significant statistical correlation with CAM scores.
Oxygen saturation <93% has been considered as an independent component in the diagnosis of severe falciparum malaria. In the study by A involving 381 children with severe falciparum malaria, hypoxia was found to be a major determinant of mortality. In the current study, hypoxia was taken as the indicator of severe malaria requiring ICU admission. Hypoxia could be caused due to the uninhibited growth of the parasites, which leads to the blocking of blood supply to vital organs-particularly, lungs-causing severe pulmonary disease [24, 25].
In the present study, there were no deaths among the patients with CAM scores <2. The sensitivity and specificity of the CAM score in identifying patients who required ICU care was 100% and 86.04%, respectively, with hypoxia as the proxy criteria for ICU admission. A PPV of 73.9% and negative predictive value of 100% denoted that the predictive value of survival in patients with low CAM scores was accurate rather than assessing the morbidity and mortality events in patients with CAM scores ≥2. Our study results are in accordance with the study of Hanson et al. ; in that study, out of the 789 patients in SEAQUAMAT for whom the CAM score was calculated, 206 (26.10%) patients succumbed to death. Out of these, only 8 patients had a low CAM score, indicating the mortality events increased with the CAM scores. Only 8 (3.6%) patients died (PPV for survival, 96.4%; 95% CI, 93.1–98.5) out of the 225 patients with low CAM scores. Further, 6 (75%) patients among these 8 patients had renal failure; and only 1 patient out of these 6 patients was able to receive dialysis. Pregnancy was a risk factor for 1 patient among the 2 patients without renal failure; there were no additional risk factors identified during hospital admission for the other patients . Similar results were obtained in the validation series conducted in Bangladesh and Vietnam [10, 11]. In the study by Marks et al. , the PPV for survival in patients with CAM scores <2 was 100% and a CAM score ≥2 was significantly associated with an increased risk of death (p=0.02).
In our study, the sensitivity was 100% for both BCAM and RCAM scores, and the specificity differed with the value of 81.39% and 90.69% for BCAM and RCAM scores, respectively. The PPVs were 68% and 80.95% for the BCAM and RCAM scores, respectively, which signified that the RCAM score was better than BCAM score in predicting morbidity and mortality events in patients. The negative predictive value was 100% for both BCAM and RCAM scores as they could accurately predict a good prognosis in patients with low scores. Overall, for the prediction of morbidity and mortality events in patients with high CAM scores, the RCAM score was found to be precise and more valuable than the other scores. In the study by Newton et al.  regarding prognostic indicators in adults hospitalized with severe falciparum malaria, BCAM and RCAM scores were calculated for 662 and 873 patients, respectively. The mortality rate increased at higher scores. Out of the 498 patients with a low BCAM score (<2), 3 (0.6%) patients died (PPV (95% CI) for survival, 99.4 (98.2%–99.8%), and out of the 675 patients with low RCAM scores (<2), 11 (1.6%) patients died (PPV for survival was 98.4 (97.1%–99.1%). The optimal cut-off value in this data set for both BCAM and RCAM was <2, with sensitivity and specificity of 93.8% and 80.6% for BCAM and 81% and 81% for RCAM, respectively .
Stepwise logistic regression analysis to find the association of CAM scores with outcome events had a high statistical significance (p<0.001). The need of ICU care increased to 16 times, 21 times, and 12 times in patients having CAM, BCAM, and RCAM scores ≥2, respectively. This clearly depicts that CAM scores ≥2 have a high prediction rate of morbidity in patients having severe malaria and should alert the treating physician to immediately shift the patient to an ICU facility in order to decrease fatal outcomes.
The limitations of our study are the following. 1) Our study was conducted in adult patients who suffered from severe malaria. Even though acidosis, respiratory distress, and altered consciousness predict a prognosis in children suffering from severe malaria, its validation in the pediatric population is yet to be proven by further studies. 2) Independent indications for ICU admission, such as refractory shock or persistent hypoglycemia may be of concern in patients with severe malaria; however, in the present study, we did not encounter indications without including the variables of CAM scores. 3) The study has been conducted in low-transmission areas of India. Further studies in high-transmission areas require the validation of CAM scores. 4) The study extends the predictive value of the CAM scores in patients with severe vivax malaria. As the number of patients with vivax malaria is small and this is the first study to extend the usefulness of CAM scores regarding the prognostication of vivax malaria, supportive studies in a large number of patients are needed to strengthen its value.
The simplicity and effectiveness of CAM scores have dramatically high influence in classifying patients with severe malaria for the effective utilization of the intensive care facility. Although many areas of India have resource-poor hospital settings, the strength of the scoring system would direct the physicians in prompt referral to the better centers with well-equipped emergency care facilities. In densely populated and developing countries like India, the genuine utility of CAM scores in severe malaria could make an enormous difference in the management of the disease and prevent highly fatal complications occurring in the course of the disease. There is an emerging need to use the scoring system by healthcare providers to give the maximum benefit to patients.
Ethics Committee Approval: Ethics committee approval was received for this study from the ethics committee of Pt B.D. Sharma University (UHS/Rtk/2014/Et-I/13).
Informed Consent: Written informed consent was obtained from patients who participated in this study.
Peer-review: Externally peer-reviewed.
Author contributions: Concept - H.K.A., D.J.; Design -D.J., A.R.; Supervision - H.K.A., R.K.; Resource - H.K.A., R.K.; Materials - D.J., A.R.; Data Collection and/or Processing - A.R., D.J.; Analysis and /or Interpretation -A.R., R.K.; Literature Search - A.R., D.J.; Writing - A.R., D.J.; Critical Reviews - H.K.A., R.K.
Conflict of Interest: No conflict of interest was declared by the authors.
Financial Disclosure: The authors declared that this study has received no financial support.