Pre-hospital oily chloramphenicol treatment in meningitis outbreaks in low resources environments

Introduction

Bacterial meningitis is widespread in the meningitis belt of sub-Saharan Africa and still has a high case fatality, especially in major epidemics in rural areas, where health facilities are distant. ¹

Long-acting oily chloramphenicol (CAP) suspension has been recommended by the World Health Organization (WHO) since 1995 ² and still remains active against N. meningitidis. ³ Alternative antimicrobial drugs with proven efficacy, such as benzyl-penicillin, ampicillin and ceftriaxone, are also suggested, ⁴ but protocols using multiple injections are impractical for use in epidemics, ⁴ while ceftriaxone is often unavailable and much too expensive in low- and middle-income countries (LMICs).

A recent systematic review failed to demonstrate effectiveness and safety of pre-admission antibiotics versus no treatment in the absence of randomised controlled trials. ⁵

A comparison trial between pre-treatment versus no treatment is unacceptable for obvious ethical reasons and only trials comparing different antibiotic treatment are available in the literature.

Benzylpenicillin given before referral to hospital has proven to be effective in reducing the mortality rate, ⁶ but resistance to this drug is increasing ⁷ and it has a short half-life. Both oily CAP and ceftriaxone have a long half-life (24 h and 48 h, respectively) and can be used for pre-treatment.

During the 2001 meningitis epidemic in Goundi sanitary district, Moyen Chari, south Chad, a high mortality rate in populations living far from the hospital suggested the need for pre-admission treatment to prevent early deaths and allow patients to survive the long journey to the hospital. ⁸

We present comparative data on pre-admission versus hospital oily CAP treatment.

Methods

Goundi district covers a population of about 121,000 individuals in a 45-km diameter region where seven health centres provide first-line care for most uncomplicated pathologies.

Intra-hospital treatment of choice for meningitis in Goundi sanitary district was CAP, according to WHO recommendations, ² considering the fact that ceftriaxone was not available in Chad in 2001.

During the meningitis outbreak, health centres far from the hospital were equipped with oily CAP (AB-Chloramphenicol 500, oil suspension, ASTRAPIN pharma GmbH & Co, Germany) and nurses were trained to recognise symptoms of meningitis and to refer patients to hospital after the first CAP administration.

Eighty-six patients treated in extra-hospital points of care, in whom meningitis was subsequently confirmed at hospital arrival, were enrolled in the study.

In the control group, we considered meningitis patients going directly to hospital and those coming from close health centres, for whom pre-treatment was not authorised.

We collected data recording lethality rate, failure of treatment and mean hospital stay from both groups, comparing cases treated in the same period, ⁹ trying to avoid bias owing to epidemiological evolution during the outbreak, such as involvement of less fragile patients and microbiological changes in bacterial behaviour.

Statistical significance was calculated with χ² and the Student’s t-test. A P value of < 0.05 was considered statistically significant.

All participants provided verbal informed consent before being enrolled.

All clinical data related to the 2001 meningitis epidemic are written in medical records stored in Goundi hospital archives.

Results

It was not possible to evaluate the number of prevented deaths in pre-treated patients, but comparison with patients receiving CAP at hospital admission showed a statistically significant reduction in pre-treated patients in mean hospital stay, lethality rate and need of second line therapy or adjunctive antibiotics (Table 1).

OPEN IN VIEWER

Table 1.

Treatment, response and statistical significance in pre-treated (p-CAP) and hospital treated patients (H-CAP).

	Patients	Deaths (lethality)	Failed first line	Mean hospital stay (days)
H-CAP	168	15 (8.9%)	18 (10.7%)	5.06
p-CAP	86	1 (1.15%)	1 (1.1%)	4.38
P value		0.02	0.01	< 0.001

Discussion

Bacterial meningitis still remains an important cause of death and disability in sub-Saharan Africa and its clinical management is among the most vital current medical challenges. ¹⁰ While vaccination programs are still necessary and effective, ¹¹ they need optimisation to improve their impact during epidemic outbreaks, when delayed administration and fast availability are a limiting factor in efficacy.^8,12

For this reason, new clinical strategies are strongly needed to face epidemic emergencies, in which pre-hospital treatment practice could offer a rapid medical intervention and overcome limitations owing to distance from hospital facilities, especially in rural areas in LMICs. ⁵

During the 2001 meningitis epidemic in south Chad, 595 cases were recorded in Goundi sanitary district hospital and the overall lethality was 8.74%, ⁸ while in other hospitals of south Chad, lethality approached 20%, although official data were not available. In all peripheral health centres, nurses were instructed and authorised to administrate oily CAP in suspected meningitis cases.

Treatment cost is one of the most important problem in rural contexts and for this reason ideal dosage of antibiotic treatment is seldom achieved: oily CAP was economically sustainable in Goundi sanitary district (the cost of ten days antibiotic treatment at official Chadian drug prices was €17.44) ⁸ and it was easily available and practical to use.

Comparison with patients treated with CAP directly in hospital showed a significant reduction of mortality, time of hospitalisation and need of second line treatment.

Data concerning patients who received pre-treatment in health centres but did not access a hospital were not considered, but we speculate oily CAP had very likely prevented some extra-hospital deaths. We suppose that some treated patients not coming to hospital could have benefited from the treatment administered in the health centre, while we acknowledge that it is not possible to gather accurate statistics of deaths during the referral journey.

Another limit of our study is the deficiency of information about over-diagnosed meningitis cases in health centres, but the risk of hypothetically unnecessary treatment is balanced by overtly demonstrated benefits. It would be advisable to invest resources on nurses' training and skilling, especially in rural contexts, overcoming the lack of doctors.

In LMICs, communicable diseases still cause millions of deaths each year and despite technical and financial resources provided in the last decades, epidemic outbreaks continue to occur. ¹³ Public health authorities must face the challenges of lack of facilities and resources and the unknown course any outbreaks may take.

The statistical significance of our results is strong and leads us to the conclusion that pre-hospital treatment is mandatory in LMICs, especially where people affected far from referral hospitals frequently cannot afford costs related both to hospital transfer and admission. In these circumstances, functional and oily CAP-equipped health centres will impact more on public health than data collected in hospitals can demonstrate, particularly during epidemics. It is therefore advisable to invest resources on nurse training, especially in rural context, in the recognition and pre-treatment of meningitis.