|Year : 2017 | Volume
| Issue : 1 | Page : 26-32
A simple mortality prognostic scoring system for burns
Pawan Agarwal, Sudhir Adalti, Vikesh Agrawal, Dhananjaya Sharma
Plastic Surgery Unit, Department of Surgery, N.S.C.B. Government Medical College, Jabalpur, Madhya Pradesh, India
|Date of Web Publication||13-Dec-2017|
Dr Pawan Agarwal
292/293 Napier Town, Jabalpur 482001, Madhya Pradesh
Source of Support: None, Conflict of Interest: None
Background: Several complex prognostic scoring systems are available for burn patients incorporating sophisticated investigations and use of global scales involved in the management of patients in intensive care unit. We constructed and validated a simplified scoring system for burn patients, which can be easily used in developing countries.
Materials and Methods: One hundred and eighty-two consecutive patients with burns undergoing treatment at a teaching hospital in Central India were studied prospectively. Multiple logistic regressions were used to assess the predictive power of each prognostic variable. A simple scoring system was constructed using the four most powerful, but easy to calculate, prognostic factors. This system was then prospectively validated in the next 122 consecutive patients.
Results: On multivariate analysis, total body surface area, percentage full thickness burn area, presence of inhalation burn, and serum creatinine were found significant predictors of mortality. Score was constructed using logit model using these four factors, which ranged from 4 to 20. Score correlated well with mortality; which increased with rising score. The mean score in survivors was significantly less than that in non-survivors (9.44 vs. 15.75; P < 0.0001). Cut off value of score ≥12 was associated with significantly higher mortality. The predicted and observed outcomes matched well.
Conclusion: The Jabalpur prognostic scoring system for burns is effective for prognostication in selected group of patients with burn injuries. It is simple and user-friendly because it uses only four routinely documented clinical risk factors.
Keywords: Jabalpur scoring system, mortality, prognosis, risk stratification, thermal burn
|How to cite this article:|
Agarwal P, Adalti S, Agrawal V, Sharma D. A simple mortality prognostic scoring system for burns. Indian J Burns 2017;25:26-32
| Introduction|| |
Many prognostic scoring systems are available for burn patients. These scoring systems allow stratification of patients according to severity, help in identification of patients at high risk and provide prognostic information. These range from simple rule of the thumb to those incorporating more sophisticated investigations and use of global scales involved in the management of patients in intensive care unit (ICU).,,,,,,, ICU care and sophisticated investigations may not be available for most burn patients due to infra-structural constraints in developing countries. Additionally, any scoring system aspiring to predict risks must take into consideration geographical variations, including treatment facilities, inherent in a cohort of patients. We, therefore, felt the need for a simple prognostic scoring system, which can be used easily in developing countries.
| Material and methods|| |
This prospective study was conducted in a plastic surgery unit, in a government medical college in Central India. Ethical approval from the Institutional Review Board (No. IEC/2010/11) was obtained.
This study included 304 thermal burn patients over a period of 3 years. First 182 patients seen over first 2 years were prospectively studied to analyze risk factors for mortality and to develop the prognostic scoring system (called developing group). The next 122 patients seen over the next year were used for prospectively validating the score (called validating group).
All thermal burn patients presenting within 24 h after burn were included in the study. Patients excluded from the present study were those with chemical or electric burn, those presenting to our hospital after 24 h of burn, patients transferred from elsewhere for acute care and those with thermal burn with associated trauma. The end point of the study was death or discharge of the patients.
At the time of admission, patient’s relevant data such as age, history of ignition of clothes, heart rate, respiratory rate, and temperature and associated co-morbid illnesses such as diabetes mellitus were noted. Blood samples for laboratory investigation were sent prior to the start of treatment for hematocrit, white blood cells (WBC) count, blood urea, serum electrolytes, and serum creatinine. At the time of admission, the total body surface area (TBSA) of burn was calculated using “rule of nine” in adults and with the help of “Lund and Browder chart” in children. Site and depth of burn were assessed by surface characteristics. Inhalation burn was diagnosed clinically if there was a history of burn sustained in closed space, the presence of facial burn, burnt nasal vibrissae, exposure to heavy smoke, presence of carbonaceous sputum, respiratory strider, hoarseness of the voice and tachypnoea rhonchi, rales, wheeze, and the use of accessory muscles of respiration. Bronchoscopy and Xenon 133 lung scan were not taken due to non-availability. The initial assessment was performed, and subsequent burn treatment was taken by using American Burn Association Practice Guidelines Burn Shock Resuscitation.
All the burns including major burns were resuscitated using ringer lactate solution 3 ml/kg/% of burn. Burn wound was treated with 1% topical silver sulphadiazine and systemic antibiotics when indicated according to culture and sensitivity report. Urine output was recorded after 24 h. Wound care included hydrotherapy, de-sloughing, escharectomies and debridement. Nutritional support included daily intake of protein and calories according to the basal energy expenditure. Inhalation injury was treated by nasal oxygen, bronchodilators and mucolytic agents. None of our patients received ventilatory support due to non-availability of burns ICU, and none underwent early excision and skin grafting.
The predictive factors considered for developing the score were TBSA, total full thickness burn (FTB), clothes ignition, high-risk site of burn (presence of head, neck, and perineal burns), temperature, age, associated comorbidity, Glasgow coma scale, inhalation burn, respiratory rate, urine output after 24 h, blood urea, serum creatinine, serum electrolytes, heart rate, hematocrit and WBC count. The developing data was analyzed using univariate and subsequently multivariate analyses for the identification of predictive factors of mortality. Using logistic regression analysis and the logit model approach of predictive score estimation; factor values (ranging 1–5) were assigned to these four predictive factors and the Jabalpur prognostic score for burn (JPS-B) was developed [Table 1]. JPS-B ranged from 4 to 20. Receiver operating characteristic (ROC) analysis was used to calculate accuracy and cutoff value of score. The statistical analysis was performed using the Statistical Package for the Social Sciences version 22.0 software (SPSS Inc., Chicago, IL, United States) and Medcalc, Ostend, Belgium (online free trial version).
| Results|| |
On univariate analysis of developing data, following factors were found statistically significant: TBSA (P = 0.0002), FTB (P = 0.00025), ignition of clothes (P = 0.0012), site of burn (P = 0.0013), age (P = 0.0014), sex (P = 0.0017), serum creatinine (P = 0.00012), inhalation burn (P = 0.00001), associated conditions (P = 0.00012) and urine output (P = 0.0012) and were significantly associated with mortality. However, multivariate analysis on same data revealed four statistically significant variables predicting the mortality, that is, TBSA (P = 0.00016), FTB (P = 0.00011), inhalation burn (P = 0.00056) and serum creatinine (P = 0.00010). A cutoff score of ≥12 was identified using ROC analysis which suggested a score value above which significant increase in mortality was observed. The sensitivity was 85.23%, specificity was 91.49% and accuracy was 95% [area under the curve (AUC) 0.95; standard error (SE) ±0.0144; 95% confidence interval (CI) 0.908–0.977; P < 0.0001; [Figure 1].
The validation group (122 patients) was analyzed for statistical validation of Jabalpur burn score (JBS) with respect to score’s accuracy to predict mortality.
The distribution of various predictive factors in two groups is shown in [Table 2]. Presence of FTB%, inhalation burn, % ignition of clothes, high-risk site of burn, and associated comorbidity was significantly higher in the validation group (P < 0.05). This reflected in higher mortality in the validation group (62.29% as against 48.35%), as well as higher mortality seen in subgroup 2 [Table 3] and [Figure 2]. It was observed that the mortality increased with increasing scores [Table 3], [Figure 2].
|Table 2: Comparison of predictive factors in developing and validation groups|
Click here to view
|Table 3: Correlation of JPS-B with mortality in the developing and validation groups|
Click here to view
The statistical significance of identified predictive factors and cut-off score value ≥12 was again proven when analysis of validation group was performed and compared with data of developing group [Table 4]. It was found that all four predictive factors were significantly higher (P < 0.0001) in non-survivors, in both the groups.
|Table 4: Outcome correlation with statistically identified predictive factors and JPS-B values in the developing and validation groups|
Click here to view
JPS-B performed well on prospective validation [Figure 2]. The calculated score ranged from 4 to 20 and correlated well with mortality in developing and validation groups (Spearman’s rho value of +0.81604 and +0.7705 respectively, both P = 0). The percentage of patients having score ≥12 was higher (P < 0.0001) in non-survivors in both groups. The mean score was significantly higher (P < 0.0001) in non-survivors in both groups. On ROC analysis, its sensitivity was 86.84%, specificity was 93.48%, and accuracy was 95.4% (AUC 0.954 ± 0.0182; 95%CI 0.901–0.984; P < 0.0001; [Figure 3]).
| Discussion|| |
Prognostication, that is, the likely course of a medical condition, is an important part of the management of any disease process. That is why prognostic scoring systems available for burn patients serve a very important function: assessing the severity of condition and its likely course thereby allowing stratification of risk − numerically and scientifically. This confirms one of the basic tenets of science laid down by Lord Kelvin − If you cannot measure it, then it is not science; and you cannot improve it. All scoring systems − regardless of their field − use the most predictive factors to calculate and produce an expected likelihood of outcome for a given patient, although the relative weighting of these factors varies between scoring systems.,,,
The credit for developing the first burn score goes to Weidenfeld, who in 1902 correlated TBSA and age with the mortality of his patient population. The prognostic powers of these two parameters were confirmed by the pioneering work of Bull and Squire in 1949 and later popularized by Baux in 1963 as “Baux score.”, Baux score continues to provide a good indication of the risk of mortality.,,,
Since the advent of Baux score, better understanding of pathophysiology of burn has taught the scientists the implications of FTB, inhalational injury and co-morbid illness associated with burn, and their additions have led to better prognostication models.,,
Age, TBSA, FTB, inhalational injury, and co-morbid conditions are the major determinants of prognostication in burns. TBSA and FTB were the strongest prognostic factors in our study as well. The prognostic effects of age of patient and presence of co-morbid illness are easy to understand as they both affect the outcome in study of any disease and burn is no exception.,,
However, contrary to other studies, age did not prove to be significant factor in our study. This can be explained as most of our patients were young, in the age group of 15–44 years; mean age of survivors was comparable to mean age of those who died in both developing and validation groups (28.38 ± 10.84 vs. 26.10 ± 8.55, P > 0.05 and 25.10 ± 11.11 vs. 27.23 ± 10.26, P > 0.05, respectively). This particular observation on age of our cohort also explains why pre-existing comorbidities were found in small percentage of patients in our study [Table 2].
Female gender has been shown as a risk factor for mortality by some authors, but this was not seen in our study. 73.35% of our patients were females, because most burn injuries occur as kitchen accidents as flame burns, an observation which is consistent with epidemiological studies from India.
Many studies have shown that the risk of mortality is increased by 400% if inhalational injury is present; as seen in the present study as well. Another significance of inhalation injuries lies in the fact that major strides have been made in burn care, but similar success has not been achieved with inhalation injuries. Recently, need for mechanical ventilation has been identified as a risk factor for mortality in burns patients with inhalation injury.
In addition, patients with inhalation injuries are exposed to 40–70% risk of infectious complications like pneumonia which is a major cause of morbidity and mortality in burn patients.
Endoscopic criteria have been standardized to grade the mucosal injury but non-availability of bronchoscopy prevented us from more objectively noting this parameter. Resource-constraints prevented us from analyzing inhalational burns more objectively with these endoscopic criteria.
Cause of mortality in most patients is sepsis induced multi-organ failure. This calls for dynamic scoring of global scales (APACHE II, APACHE III, SAPS, etc.) in an ICU setting.,,
These global scales, while delivering excellent prognostication, present significant logistic challenges in developing countries. Most government hospitals do not have burn ICUs/ventilatory support and are unable to perform expansive biochemical tests like blood gas analysis. This calls for simplifying scoring systems and surgical audit systems for use in developing countries with limited resources.,,,,
Burn patients are prone to develop infections; prompting the rationale of use of inflammatory response/pre-sepsis/sepsis biomarkers like serum lactate, procalcitonin, plasma TNF-α, and interleukin 8/10, etc.,, These can help in prognostication as they are found to be associated with burn injury severity. Presence of early systemic inflammatory response syndrome, and its biochemical clues have been recently shown to have adverse relationship with survival. To reiterate, most government hospitals are unable to perform expansive assays of these biomarkers.
Methodological challenges in designing and validating scores that are applicable to a wide range of populations are well known.,
Differences in patient sub-sets and standards of burn care in various places demand that any prognostic model is first validated before their application in a new population. And this is the reason why independent/ different scoring systems – delivering good prognostication, discrimination and calibration − are needed for independent population and data sets to identify the ones best suited., This prompted us to fulfill our need by analyzing our patients’ data and developing JPS-B.
However, there are a few limitations of our study. Small numbers of patients have been studied, all of which were managed conservatively and extremes of age were not seen in our study. JPS-B, though applicable in similar patient population, may require validation/addition of extra factors before its application in a different population of patient sub-sets and standards of burn care. It is well known that all conventional disease severity scoring systems are better at predicting group mortality but are not so reliable in individuals and should not be used to make decisions to deny or terminate treatment in individuals.,Additionally, physicians using such scoring systems should be alert for changes in the patient populations change and newer treatment modalities, which will require recalibration and updating of scoring system.
Qualities of a good prognostic system are well known: it should be simple and inexpensive, routinely available, reliable (intra and inter-observer), objective (observer independent), specific to the function of the organ in question, therapy independent, sequential (available at admission and then at fixed periods of time), not affected by transient, reversible abnormalities associated with therapeutic or practical interventions, reflect acute dysfunction of the organ in question but not chronic dysfunction, reproducible in large, and heterogeneous groups of patients.,,
In conclusion, JPS-B has most of these qualities and is a simple, accurate system for objectively estimating the probability of death in our burn unit using only four parameters which can be noted simply and objectively; the score can be calculated at bed side in the smallest hospital in similar age group of patients and with similar treatment modalities in a developing country, where burn-ICU facilities and sophisticated investigations are lacking.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Bull JP, Squire JR. A study of mortality in a burns unit: Standards for the evaluation of alternative methods of treatment. Ann Surg 1949;130:160-73.
Tobiasen J, Hiebert JM, Edlich RF. The abbreviated burn severity index. Ann Emerg Med 1982;11:260-2.
Ryan RM, Schoenfeld DA, Thorpe WP, Sheridan RL. Objective estimates of the probability of death from burn injuries. N Engl J Med 1998;338:362-6.
Belgian Outcome in Burn Injury Study Group. Development and validation of a model for prediction of mortality in patients with acute burn injury. Br J Surg 2009;96:111-7.
Sheppard NN, Hemington-Gorse S, Shelley OP, Philp B, Dziewulski P. Prognostic scoring systems in burns: A review. Burns 2011;37:1288-95.
Emara S. Prognostic indicators in acute burned patients − A review. J Acute Dis 2015;4:85-90.
Pantet O, Faouzi M, Brusselaers N, Vernay A, Berger MM. Comparison of mortality prediction models and validation of SAPS II in critically ill burns patients. Ann Burns Fire Disasters 2016;29:123-9.
Moore EC, Pilcher DV, Bailey MJ, Stephens H, Cleland H. The Burns Evaluation and Mortality Study (BEAMS): Predicting deaths in Australian and New Zealand burn patients admitted to intensive care with burns. J Trauma Acute Care Surg 2013;75:298-303.
Sharma D. Scoring systems for intra-abdominal sepsis: The science of predicting risks. Indian J Surg 2002;64:125-47.
Pham TN, Cancio LC, Gibran NS. American Burn Association practice guidelines burn shock resuscitation. J Burn Care Res 2008;29:257-66.
Baux S. Contribution to the % use of local treatment of extended thermal burns. Thesis. Paris, France; 1961.
Roberts G, Lloyd M, Parker M, Martin R, Philp B, Shelley O et al.
The Baux score is dead. Long live the Baux score: A 27-year retrospective cohort study of mortality at a regional burns service. J Trauma Acute Care Surg 2012;72:251-6.
Karimi H, Motevalian SA, Rabbani A, Motabar AR, Vasigh M, Sabzeparvar M et al.
Prediction of mortality in pediatric burn injuries: R-Baux score to be applied in children (pediatrics-Baux score). Iran J Pediatr 2013;23:165-70.
Williams DJ, Walker JD. A nomogram for calculation of the Revised Baux Score. Burns 2015;41:85-90.
Steinvall I, Elmasry M, Fredrikson M, Sjoberg F. Standardised mortality ratio based on the sum of age and percentage total body surface area burned is an adequate quality indicator in burn care: An exploratory review. Burns 2016;42:28-40.
Gomez M, Wong DT, Stewart TE, Redelmeier DA, Fish JS. The FLAMES score accurately predicts mortality risk in burn patients. J Trauma 2008;65:636-45.
Osler T, Glance LG, Hosmer DW. Simplified estimates of the probability of death after burn injuries: Extending and updating the Baux score. J Trauma 2010;68:690-7.
Moore EC, Pilcher DV, Bailey MJ, Cleland H, McNamee J. A simple tool for mortality prediction in burns patients: APACHE III score and FTSA. Burns 2010;36:1086-91.
Moore EC, Pilcher DV, Bailey MJ, Stephens H, Cleland H. The burns evaluation and mortality study (BEAMS): Predicting deaths in Australian and New Zealand burn patients admitted to intensive care with burns. J Trauma Acute Care Surg 2013;75:298-303.
Dokter J, Meijs J, Oen IM, van Baar ME, van der Vlies CH, Boxma H. External validation of the revised Baux score for the prediction of mortality in patients with acute burn injury. J Trauma Acute Care Surg 2014;76:840-5.
Heng JS, Clancy O, Atkins J, Leon-Villapalos J, Williams AJ, Keays R et al.
Revised Baux score and updated Charlson comorbidity index are independently associated with mortality in burns intensive care patients. Burns 2015;41:1420-7.
Knowlin L, Stanford L, Moore D, Cairns B, Charles A. The measured effect magnitude of co-morbidities on burn injury mortality. Burns 2016;42:1433-8.
Pompermaier L, Steinvall I, Elmasry M, Thorfinn J, Sjöberg F. Burned patients who die from causes other than the burn affect the model used to predict mortality: A national exploratory study. Burns 2017. doi: 10.1016/j.burns.2017.07.014
Goswami P, Singodia P, Sinha AK, Tudu T. Five-year epidemiological study of burn patients admitted in burns care unit, Tata Main Hospital, Jamshedpur, Jharkhand, India. Indian J Burns 2016;24:41-6. [Full text]
Lionelli GT, Pickus EJ, Beckum OK, Decoursey RL, Korentager RA. A three decade analysis of factors affecting burn mortality in the elderly. Burns 2005;31:958-63.
Kim Y, Kym D, Hur J, Yoon J, Yim H, Cho YS et al.
Does inhalation injury predict mortality in burns patients or require redefinition? PLoS One 2017;12:e0185195. doi: 10.1371/journal.pone.0185195
Trupkovic T, Gille J, Fischer H, Kleinschmidt S. [Antimicrobial treatment in burn injury patients]. Anaesthesist 2012;61:249-51, 254-6, 258.
Ikonomidis C, Lang F, Radu A, Berger MM. Standardizing the diagnosis of inhalation injury using a descriptive score based on mucosal injury criteria. Burns 2012;38:513-9.
Dokter J, Felix M, Krijnen P, Vloemans JF, Baar ME, Tuinebreijer WE et al.
Mortality and causes of death of Dutch burn patients during the period 2006–2011. Burns 2015;41:235-40.
Douglas HE, Ratcliffe A, Sandhu R, Anwar U. Comparison of mortality prediction models in burns ICU patients in Pinderfields Hospital over 3 years. Burns 2015;41:49-52.
Agrawal S, Sharma D, Raina VK. Arterial pH and arterial oxygenation are not essential for risk stratification in perforation peritonitis. Indian J Gastroenterol 1999;18:5-6.
Agrawal V, Sharma D, Raina VK. Correlation of trauma scoring systems with abdominal septic complications in abdominal trauma. Indian J Gastroenterol 2002;21:188-92.
Mishra A, Sharma D, Raina VK. A simplified prognostic scoring system for peptic ulcer perforation in developing countries. Indian J Gastroenterol 2003;22:49-53.
] [Full text]
Parihar V, Sharma D, Kohli R, Sharma DB. Risk adjustment for audit of low risk general surgical patients by Jabalpur-POSSUM score. Indian J Surg 2005;67:38-42.
Singh H, Mishra A, Sharma D, Somashekar U. A simple prognostic scoring system for typhoid ileal perforation peritonitis. Trop Doct 2010;40:203-7.
Tsurumi A, Que YA, Ryan CM, Tompkins RG, Rahme LG. TNF-α/IL-10 ratio correlates with burn severity and may serve as a risk predictor of increased susceptibility to infections. Front Public Health 2016;4:216.
Lunawat A, Vashistha RK, Patel V, Chhabra R, Kolla V. Predicting mortality in burns: A new scoring system. Int Surg J 2016;3:271-6.
Brusselaers N, Agbenorku P, Hoyte-Williams PE. Assessment of mortality prediction models in a Ghanaian burn population. Burns 2013;39:997-1003.
Hussain A, Choukairi F, Dunn K. Predicting survival in thermal injury: A systematic review of methodology of composite prediction models. Burns 2013;39:835-50.
Salehi SH, As’adi K, Abbaszadeh-Kasbi A, Isfeedvajani MS, Khodaei N. Comparison of six outcome prediction models in an adult burn population in a developing country. Ann Burns Fire Disasters 2017;30:13-7.
Tsurumi A, Que YA, Yan S, Tompkins RG, Rahme LG, Ryan CM. Do standard burn mortality formulae work on a population of severely burned children and adults? Burns 2015;41:935-45.
Rogers J, Fuller HD. Use of daily acute physiology and chronic health evaluation (APACHE) II scores to predict individual patient survival rate. Crit Care Med 1994;22:1402-5.
Turner JS, Potgieter PD, Linton DM. Systems for scoring severity of illness in intensive care. S Afr Med J 1989;76:17-20.
Marik PE, Varon J. Severity scoring and outcome assessment. Computerized predictive models and scoring systems. Crit Care Clin 1999;15:633-46.
Marshall JC. Multiple organ dysfunction syndrome. In: Sibbald WJ, Vincent JL, editors. Clinical Trials for the Treatment of Sepsis. Heidelberg: Springer-Verlag; 1995. p. 122-38.
Vincent JL, Moreno R. Clinical review: Scoring systems in the critically ill. Crit Care 2010;14:207.
[Figure 1], [Figure 2], [Figure 3]
[Table 1], [Table 2], [Table 3], [Table 4]