Neonatal sepsis in Peshawar

Author’s reply

I am responding to the letter of Ali et al.¹

(1) The total number of babies (1598) and number of culture positive babies (1003) in our article² represent the number of cases after removal of patients meeting the exclusion criteria. We did find isolated cases of Streptococcus sp, Salmonella, and Enterococci, but they happened to fall in the excluded group.

These organisms accounted for < 0.5% of the spectrum as a whole. This is similar to the findings of Maryam et al³ in a public sector institution with a population of similar socioeconomic, cultural, religious, and climatic background. Their study was carried out in the same time period as ours but completely independently and blinded from ours. In their series of 284 cases, they grew Escherichia coli (130 cases; 45.7%), Klebsiella (49 cases; 17.2%), Pseudomonas (46 cases; 16.2%), Staphylococcus aureus (39 cases; 13.7%), Staphylococcus epidermis (18 cases; 6.34%), Strepococcus sp (1 case; 0.3%), and Salmonella (1 case; 0.3%).

The numbers are slightly different among the studies from south of Pakistan⁴. This is not surprising as neonatal sepsis is known for the temporal and regional variation of the spectrum of its organisms even in different hospitals within the same city.

(2) The basic message from the majority of studies from Pakistan is the same: Gram negative organisms are the main cause of neonatal sepsis in Pakistan, followed by S aureus. This group of organisms is responsible for > 99% of the spectrum, and unfortunately the grave situation of multidrug resistance is emerging among these organisms. That is where research needs to be concentrated, instead of on the organisms responsible for < 0.5% of the spectrum (Salmonella, Streptococcus sp, etc), which do not carry any significance for overall neonatal mortality and morbidity.

(3) Out of 296 cases of S aureus in our series, ampicillin was tested on 285 cases, with 171 (60%) sensitive to it, and 279 were tested with augmentin, with 75 (26.9%) sensitive to it. I agree with Ali et al that this pattern of sensitivity looks unusual as far as S aureus is concerned, although this phenomenon is known to occur with β lactamase-producing E coli. It may be due to the various strengths of augmentin discs available, the known biochemical instability of clavulonic acid, or the difficulty of interpretation when a combination of two antibiotics is used in one disc using the disc diffusion technique. However, I would be interested to hear more expert opinion on this. Our series did not exclude hospital acquired infections.

(4) The longitudinal analysis of our data shows an increasing sensitivity to penicillin and decreasing sensitivity to cephalosporins, particularly cefotaxime, over the last half decade. This is consistent with the change in antibiotic use in Pakistan since the early 1990s when penicillin/gentamicin was replaced by cephalosporins/amikacin as the first line antibiotic treatment. Most of the Gram negative organisms in Pakistan maintain a very high degree of sensitivity to amikacin^3,4 but not to gentamicin. I feel that penicillin/amikacin may be a very good choice as the first line antibiotic in neonatal units in Pakistan. It is high time that we reviewed our antibiotic policies and at the same time approach the government to rationalise antibiotic marketing in this country.