Analysis Of Blood Culture Results Of Burn Septicaemia Patients Over A Period Of Nine Years In The Baghdad Burn Medical Centre

Summary

Burn is one of the most devastating traumas that someone can encounter in their life. Burn wound sepsis is still the leading cause of death in burned patients. Appropriate knowledge of the causative pathogen in burn sepsis is important for successful patient management and for the reduction of the incidence of antibiotic resistance. A retrospective study was conducted between 2010 and 2018 at the Burn Specialty Hospital in Baghdad. A total of 320 blood culture samples were obtained from patients with sepsis or suspected of having sepsis. Patient age ranged between 9 months to 70 years old, with a mean total burn surface area of 45.26%. The most common microorganisms isolated from those patients who had sepsis or suspicion of sepsis were Klebsiella (48 cases) followed by Pseudomonas (36 cases), Staphylococcus species (26 cases), Enterococcus (8 cases), Acinetobacter (11 cases), E-Coli (11 cases), Candida (4 cases), Proteus (2 cases), and Salmonella, Streptococcus pneumonia, Monilia, and Seriata one case for each. The most commonly isolated organism was Klebsiella: it was sensitive to Imipenem followed by Amikacin, Nitrofurantoin, Piperacillin, Ciprofloxacin, Co-trimoxazole, Chloramphenicol, Tetracycline, Azithromycin and Cefotaxime. Microbiological surveillance of burn patients with sepsis or suspicion of having sepsis over a period of 9 years in our hospital has shown that the most common microorganism isolated from blood cultures was Klebsiella. Klebsiella was sensitive to Imipenem mainly according to sensitivity testing using the disk diffusion method.

Introduction

Burn can be defined as a traumatic lesion that affects different layers of the skin and underlying layers. It can be caused by different factors including thermal, electrical and chemical causes. Up to 322,000 people die annually from burn worldwide according to the World Health Organization (WHO), and a large percentage of deaths (>95%) occur in developing countries. Due to improvement in burn resuscitation, wound care, nutritional support and control of infection the survival rate of burn has dramatically improved in recent years. However, 75% of deaths in burn patients are due to sepsis.^1,2,3

A few hours after the burn occurs, the surface of the burn will become contaminated with a variety of bacterial types that gradually multiply and grow. Due to impairment of the immune system of burned patients, these bacteria will seed and multiply in other parts of the body, such as the kidneys, lungs and bloodstream. The source of bacteria in the burned patient could be either normal flora (endogenous) from the patient themself or exogenous from the environment or healthcare staff. In general, exogenous organisms have more resistance than endogenous ones. The organisms that are linked to infection in burn patients could be either gram-positive or negative bacteria and could be yeast or fungi or even viruses.^4,5

Microbiological analysis using wound swabs, quantitative cultures and blood cultures may be useful to diagnose burn wound infection and sepsis. The appropriate knowledge of the causative pathogens in burn sepsis with established surveillance of the most prevalent bacteria in burn wards together with knowing its antimicrobial resistance is considered as one of the measures that successfully control infections in burn patients with reduced incidence of antibiotic-resistant bacteria emergence.^6,7,8

In this retrospective study, we investigated the common pathogenic bacteria obtained by blood culture and its antimicrobial sensitivity among patients admitted to the Burn Specialty Hospital at the Medical City Complex in Baghdad with sepsis or suspicion of sepsis over a period of nine years.

Materials and methods

This study is a retrospective study that was conducted at the Burn Specialty Hospital of the Medical City in Baghdad. The data for this study were obtained from patient medical records from the 1^st of January 2010 to December 31^st 2018. Medical records for all patients were reviewed with specific attention paid to identifying the pathogenic bacteria that were isolated from blood cultures obtained from patients who had sepsis or were suspected of having sepsis. Sepsis is defined as systemic inflammatory response syndrome (SIRS) to an infectious process. SIRS was defined as the following:

1. Body temperature either more than 38o Cor below 36^o C.

2. Heart rate more than 90/minute.

3. Respiratory rate more than 20 /minute or PaCo2 less than 32 mmHg.

4. White blood cells (WBC) more than 12000/mm³ or less than 4000/mm³.^4,9

Blood culture used in this study was processed using standard microbiological techniques. The culture samples were obtained from a peripheral vein using meticulous aseptic techniques to prevent blood culture contamination. The site of venepuncture was prepared using an antibacterial disinfectant such as 2% iodine or 0.5 chlorhexidine in 70% alcohol. The disinfectant agent was allowed to evaporate before the blood sample was withdrawn. In all cases, a reasonable quantity of blood was taken, at least 10 ml blood sample for adults, 2-5 ml for children and 1-2 ml for neonates.

The blood taken was separated into two tubes, the first tube vented for recovery of the aerobic organism’s culture, and the second tube non-vented for the anaerobic organism’s culture. In general, no anticoagulant was needed in the tube as the blood sample was immediately added to a sufficient amount of broth (50 ml). The blood culture broth used in our laboratory is a brain-heart infusion (5 ml of blood in 50 ml of broth). The blood culture bottles were then incubated at 35±1^o C and were inspected twice daily for signs of microbial growth. Growth is evident by many characters like gas production, haemolysis, uniform or subsurface turbidity, coagulation of broth... etc. When the visible growth appeared, the bottle of blood culture was opened in aseptic conditions using a sterile loop or Pasteur pipette. A small amount of broth was removed and a gram stained smear was used to examine for the presence of microorganisms. Subculture was used for specific microorganisms, for example staphylococci using blood agar or mannitol salt agar. The antimicrobial sensitivity of microorganisms was determined using the disk diffusion method.

Results

This retrospective study was conducted at the Burn Specialty Hospital in Baghdad from January 1^st, 2010 to December 31^st, 2018. A total number of 320 blood culture samples were isolated from about 320 patients. Patient age ranged between 9 months to 70 years. There were 136 males and 184 females: the main causes of burn were flame (158 patients), scald (102 patients), electrical injury (46 patients) and chemical injury (14 patients). The mean total burn surface area was 45.26%.

All these patients had a septic episode or suspected sepsis. 60% of patients had their septic episode between 3-14 days post-burn and about 40% had a septic episode between 14-21 days post-burn. Blood culture was positive in 149 specimens and no growth was noticed in 171 specimens, as shown in Fig. 1.

Fig. 1. Specimen growth.

The most common isolated microorganism was Klebsiella (48 cases), followed by pseudomonas (36 cases), Staphylococcus species (26 cases), Enterococcus (8 cases) Acinetobacter (11 cases), E-Coli (11 cases), Candida (4 cases), Proteus (2 cases) and Salmonella, Streptococcus pneumonia, Monilia, and Seriata one case each, as shown in Fig. 2.

Fig. 2. Common isolated microorganisms.

The antibiotics to which the bacteria were susceptible, as shown by the sensitivity test, were Imipenem followed by Amikacin, Nitrofurantoin, Piperacillin, Ciprofloxacin, Ceftazidime, Co-trimoxazole, Vancomycin, Chloramphenicol, Tobramycin, Ticarcillin, Tetracycline, Gentamicin, Ceftriaxone, Clindamycin, Ampicillin, Azithromycin and Cefotaxime. Isolated microorganisms from the blood cultures and sensitive antibiotic distribution in all the study cases are shown in Fig. 3 below.

Fig. 3. Microorganisms isolated from the blood cultures and the sensitive antibiotic, according to the sensitivity test distribution, in all the study cases.

The most commonly isolated organism, which was Klebsiella, was sensitive to Imipenem, Amikacin, Nitrofurantoin, Piperacillin, Ciprofloxacin, Co-trimoxazole, Chloramphenicol, Tetracycline, Azithromycin and Cefotaxime, as shown in Fig. 4 below.

Fig. 4. The most commonly isolated organism, Klebsiella, and the medications it was sensitive to.

Discussion

Burn is a major and devastating global health problem. It is especially serious in developing countries. As a result of the prolonged hospital stay and immunosuppression effect of burn, together with the direct nature of burn itself, all of these factors make patients with burns susceptible to infections. Both the depth and the size of the burn surface area, together with the patient’s age, affect the incidence of burn wound sepsis. Sepsis occurs mostly in the extreme age groups and rarely occurs in partial-thickness burns and smaller burned surface areas.

Multidrug resistance is an emergent problem in burn patients, which occurs mainly due to prolonged courses of broad-spectrum antibiotics that make treating infections in burn patients a challenging problem. Microbiological surveillance of burn patients will facilitate knowing the microorganism and the appropriate choice of antibiotic.^{6,10,11,12,13}

Positive blood culture in a patient with suspected sepsis will guide us to the appropriate treatment, and knowing the most prevalent microorganism that causes sepsis in our burn unit will facilitate the selection of appropriate and most selective antibiotics, which reduces costs and also reduces multidrug resistance. In this retrospective study, we reviewed 320 blood cultures in the period from 2010 to 2018 for burn patients who were admitted to our hospital and developed septic episodes or suspected sepsis during their treatment course in the Burn Specialty Hospital at Baghdad Medical City Complex. In this study females were affected more than males (57.5% versus 42.5%). This may be attributed to the fact that women spend more time near the fire compared to males, they have a higher incidence of self-inflicted flame burn suicide compared to men in our culture unfortunately, and they have a higher percentage of body surface area burns compared to men in our hospital due to the aforementioned reasons. Males in this study, on the other hand, have work-related burns which are mostly flash burns, contact burns, chemical burns, and electrical burns, in which a smaller body surface area is involved and therefore a shorter hospital stay is required, and there is a lower risk of burn wound sepsis development. This variation in sex is visible when we know that in spite of having more male patients treated in our hospital compared to females over the nine years (from 2010- 2018) (14.741 males – 9.663 females) the overall mortality rate in those years was higher in female patients (296 males – 364 females). This variation in sex was in agreement with Kaur et al.,¹⁴ where they investigated bacterial profile in 50 burn patients for blood and burn wound infection and they found that 68% of their patients were females and 32% of them were males. Regarding sex variation of septic episode in burn patients, our study was contradicted by the findings of Bang et al., where they observed that septicaemia occurred in 56% males and 44% females. This finding was noticed in a study that was conducted between June 1992 and May 1996 on 943 patients at the burn unit of Al-Babtain Centre for Plastics and Burns, where 79 (28%) out of 943 patients developed clinically and microbiologically confirmed septicemia.¹⁵

Our sample of patients was restricted to those who had sepsis or were suspected of having sepsis and, as we noted above, females were affected more than males. Other studies, which investigated burn wound infection in general, showed results that contradicted our finding regarding sex distribution. Gaffer et al. found burn wound infection to be more common in males (62.4) than females (37.6). Macedo and Santos also found burn wound infections to be more frequent in males than females (59.1% versus 40.9%). A similar finding was recorded by Vostrugina et al., where more males were affected by burn wound infections than females (76% versus 24%). A similar finding was also noted by Bagdomas et al.¹⁶

In our study, the majority of septic episodes happened in the first 2 weeks post-burn. This is consistent with Macedo et al.’s observations. They noted that the majority of septic episodes had developed in the first two weeks post-burn.17 Bang et al. also observed that septicaemic episodes occurred within 2 weeks post-burn (68%), with the maximum number occurring between 6-10 days post-burn.¹⁵

Flame burn was the leading cause among our patients who had septicaemic episodes (46.87%). This is consistent with Macedo et al.¹⁷ These results might be attributed to the effect of flame burn, which produces more extensive damage and more wound colonization. The mean total body surface area burned in the patients involved in this study was 45.26. The mean total body surface burned of patients with burn sepsis noted by Macedo et al. was 37.7±18.4%, ranging between 7-84%.¹⁷ Bang et al. reported that their patients with burn sepsis had a mean total surface area burned of 46% (ranging between 10-90%).¹⁵

Of the 320 blood culture samples, 171 cases (53.4%) revealed no microorganism growth, while 149 cases (46.56%) showed positive microorganism growth. In general, culture-negative sepsis could occur in 41.5% of cases, according to Phua and colleagues based on their large prospective study on patients presenting with severe sepsis.

Culture-negative sepsis patients represent an extremely challenging issue for both the clinician and the microbiologist. This may be attributed to many causes: firstly it may be due to antibiotic administration prior to blood culture sample being taken; secondly, it may be due to insufficient or incomplete diagnostic microbiological workup; lastly, it may be due to unusual organisms that cause sepsis which cannot be identified. Some attribute culture-negative sepsis to non-infection causes of sepsis.¹⁸

In our study, the major causative organism in septicaemic patients was Klebsiella, followed by Pseudomonas then by Staphylococcus species. Our results are inconsistent with Macedo et al., who observed that most septic episodes were due to the Staphylococcus species.¹⁷ Our results are also inconsistent with Bang et al., who noticed that 62% of septicaemic cases were due to gram-positive bacteria, 25% were due to gram-negative bacteria, and 13% were due to mixed organisms. Bang et al. showed that 48 episodes of septicaemia (41%) were due to Methicillin-Resistant Staphylococcus aureus (MRSA), followed by 17 (14.4%) due to Methicillin-Resistant Staphylococcus epididymis (MRSE). Bang et al. in another study confirmed that the majority of sepsis episodes in scald (82.4%) and flame burn (56.7%) were attributed to gram-positive organisms, especially MRSA.¹⁹ Sanyal et al. found that 61% of septicaemic attacks occurred due to gram-positive cocci (76% of these gram-positive cocci were MRSA), while only 26% were due to gram-negative bacilli (mainly Pseudomonas 38%, Acinetobacter 35%, followed by Coliforms in 25%).

However, our study is consistent with Dasari et al. in finding that gram-negative cocci are the major cause of septicaemic episodes in burn patients. However in their study they showed that Pseudomonas aeruginosa accounted for 31% of cases, while Klebsiella species accounted for 24% of cases. Dasari et al. showed that 11% of septicaemic episodes were due to staphylococcus aureus and 30% were due to multiple organisms. In 9% of cases no isolated organism was revealed.²⁰

In another study conducted by Sharma et al., they found that 34% of isolated organisms in septicaemic patients were Klebsiella, followed by Pseudomonas aeruginosa in 30%, which is in agreement with our study results.21 Both Evol et al. and Agnibouti et al. reported similar findings, consistent with our own.²⁰

Those reports that showed MRSA/MRSE to be the most common causes of sepsis in burn patients attributed this to the fact that the burn unit is the main source of MRSA, and they regarded MRSA to be widespread in the hospital environment. Some of the above-mentioned reports attributed low incidence of gram-negative bacilli in septicaemic burn patients to topical antibiotic use and effective systemic antibiotic.^15,17

In our study, interestingly, we observed septicaemic episodes due to Acinetobacter species.

Acinetobacter was rarely observed in burn wound sepsis until 1970. Since 1985 this organism had been isolated frequently. It was the fourth most common organism isolated from blood culture. The emergence of Acinetobacter species is attributed to the extensive use of broad-spectrum antibiotics. Acinetobacter species are usually multi-resistance bacteria.^15,17

A commonly isolated bacterium in this study, namely Klebsiella, was sensitive to Imipenem followed by multiple drugs. Both Imipenem and Amikacin proved to be effective against most of the isolated microorganisms from blood culture, as shown by sensitivity testing. Lastly, Staphylococcus species proved to be highly sensitive to Vancomycin.

In our study, Candida species were isolated from blood culture. Septicaemia due to Candida species is not uncommon in burned patients, especially those with high total burn surface area (more than 50%), prolonged usage of multiple antibiotics and in those patients with prolonged hospital stay.¹⁷

Conclusion

Microbiological surveillance of sepsis in burn patients is important to determine the causative pathogens and to choose the appropriate and effective antibiotic that reduces the chance of multidrugresistant pathogens. This study revealed that the most common microorganism that causes sepsis in our burn unit is Klebsiella. In our study Klebsiella was sensitive to Imipenem. Both Imipenem and Amikacin proved to be effective against the majority of the isolated microorganisms from blood culture, as shown by sensitivity testing using the disk diffusion method.