Isolation of klebsiella from neonatal sepsis

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

The study was conducted at the Children's Hospital and institute of Child Health during January to June 2009. A total of 12000 blood culture specimens taken from neonates were processed, out of which 1070 were positive for bacterial growth. Among these positive samples the Klebsiella spp. were 408 (38.13%) followed by Staphylococcous aureus (7.2%) and Escherichia coli (9.15%). The most effective antibiotics against Klebsiella were Pipracillin-Tazobactum (87.74%) and Ciprofloxacin (81.37%). The Amoxicillin + calvulanic acid showed good antimicrobial sensitivity pattern of 62.25%. High antimicrobial resistance was observed with Cefuroxime (95.58%), Cefixime (93.38%), Gentamycin (89.46%) and Cefetazidime (83.82%).


A total of twelve thousand blood samples were collected and processed over a time period of 6 months from January 2009 to June 2009 in the microbiology department of the Children Hospital and Institute of Child Health, Lahore.

A total of 1070 samples showed positive results. The frequency of gram positive was found to be 27.6% (296 samples) and that of gram negative was 72.3% (774 samples) (table-4.1).

Out of the pathogenic bacteria isolated from blood samples, the most common pathogen isolated was Klebsiella spp (38.13%). The next most common organism found was Staphylococcus spp (16.8%) and Escherichia coli (9.15%). Other species included Pseudomonas (8.31%), Burkholderia (6.07%), Salmonella (2.99%) etc (Table-4.2).

Antimicrobial sensitivity pattern for Klebsiella spp was noted. Klebsiella was found to be highly sensitive to Pipracillin-Tanzobactum (87.74%) and Ciprofloxacin (81.37%). Good sensitivity pattern was also observed with Amoxicillin + clavulanic acid (62.25%). While very low sensitivity was observed with Cefuroxime (4.41%), Cefixime (6.61%), Gentamycin (10.53) and Cefetazidime (16.17%) (Table-4.3).


Despite ongoing major advances in antisepsis and in the development of potent antimicrobial agents since the early twentieth century, human beings remain subject to bacterial and fungal infection through mechanisms of virulence that continue to evade the latest advents in the microbiologic field today (Linkin et al., 2004 ).

Septicemia is an infection of the blood often caused by bacteria (Jennifer, 2009). Sometimes known as blood poisoning, the bacteria in the blood release toxins than can have a serious impact on many organ systems (Marrie and Durrant, 1989).

Neonatal sepsis is invasive bacterial infection occurring in the first 90 days of life. Signs are multiple and include diminished spontaneous activity, less vigorous sucking, apnea, temperature instability, respiratory distress, vomiting, diarrhea, abdominal distention, jitteriness, seizures, and jaundice (Calderwood et al., 1993).

Neonatal sepsis occurs in 0.5 to 8.0/1000 births. The highest rates occur in low-birth-weight (LBW) infants, those with depressed respiratory function at birth, and those with maternal perinatal risk factors. The risk is greater in males (2:1) and in neonates with congenital anomalies (Lieberman et al., 1993). Declining age of mother, infection, low socio-economical status, poor and unhygienic community environments play a significant role in the chance of disease acquisition (Sturenburg et al., 2003).

Depending upon the combination of above cited factors, the disease process may manifest itself in two ways: 1) Early-onset sepsis :early-onset sepsis appears within 6 h of birth, and most cases occur within 72 Hours of life or during passage through the birth canal (Sundsfjord et al., 2004).

2) Late-onset sepsis:(after 7 days) is often nosocomial (Cao et al., 2002).

Group B Streptococcus (GBS) and gram-negative enteric organisms (predominantly Escherichia coli) account for 70% of early-onset sepsis. Vaginal or rectal cultures of women at term may show GBS colonization rates of up to 30%. At least 35% of their infants also become colonized (Spanu et al., 2002). The density of infant colonization determines the risk for invasive disease, which is 40 times higher with heavy colonization. Although only 1/100 of those colonized develop invasive disease due to GBS, > 50% of those present within the 1st 6 h of life (Wu et al., 2003). Nontypeable Haemophilus influenzae sepsis has been increasingly identified in neonates, especially premature neonates (Kollef and Fraser, 2001).

Other gram-negative enteric bacilli (e.g., Klebsiella sp) and gram-positive organisms-Listeria monocytogenes , enterococci (e.g., Enterococcus faecalis, E. faecium), group D streptococci (e.g., Streptococcus bovis ), α-hemolytic streptococci, and Staphylococci-account for most other cases. They account for 30 to 50% of late-onset cases. Streptococcus pneumoniae , H. influenzae type b, and less commonly, Neisseria meningitidis have been isolated. Asymptomatic gonorrhea occurs in 5 to 10% of pregnancies, so N. gonorrhoeae may be a pathogen. Isolation of Enterobacter cloacae or Enterobacter sakazakii from blood or CSF suggests contaminated feedings (Arredondo-Garcia, 1992).

Among all these isolates causing the neonatal sepsis Klebsiella is most important. Klebsiella is a genus of non motile, gram negative, oxidase negative, rod shaped bacteria with a prominent polysaccharide-based capsule. It can lead to a wide range of disease states, notably pneumonia, urinary tract infections, septicemia, and soft tissue infections (Daoud, 1995).

Klebsiella pneumoniae is a encapsulated, lactose fermenting, facultative anaerobic, rod shaped bacterium found in the normal flora of the mouth, skin and intestines. It is clinically the most important member of the Klebsiella genus of Enterobacteriaceae; it is closely related to K.oxytoca from which it is distinguished by being indole-negative and by its ability to grow on both melezitose and 3-hydroxybutyrate (Banerjee, 1993).

It is a common hospital-acquired pathogen, causing urinary tract infections, nosocomial pneumonia and intraabdominal infections. K. pneumoniae is also a potential community-acquired pathogen (Dobson and Baker, 1990).

K. pneumoniae has been recognized as a pulmonary pathogen since its discovery more than 100 years ago. The classic clinical presentation is dramatic: toxic presentation with sudden onset, high fever, and hemoptysis. Chest radiographic abnormalities such as bulging interlobar fissure and cavitary abscesses are prominent (Chevrolt and Pittet, 2004).

Klebsiella infections tend to occur in people with weekend immune system. Many of these infections are nosocomial. The most common infection caused by Klebsiella bacteria outside the hospital is pneumonia (Elhag, 1982). Classically, Klebsiella pneumonia causes a severe, rapid-onset illness that often causes areas of destruction in the lung (Chacko and Sohi,2005).

Infected persons generally get high fever, chills, flu-like symptoms and a cough productive of a lot of mucous. The mucous is often thick and blood tinged and has been referred to as "currant jelly" sputum due to its appearance (Gaynes and Solmon, 1996). This may involve getting blood samples, urine samples or a swab (Grandson et al., 1998).

Mortality in Klebsiella pneumonia is fairly high due to the underlying disease that tends to be present in affected persons (Higgins, 1995). While normal pneumonia frequently resolves without complication, Klebsiella pneumonia more frequently causes lung destruction and pockets of pus in the lung (known as abscesses). Klebsiella can also cause less serious respiratory infections, such as bronchitis, which is usually a hospital-acquired infection (Khan and Akram, 1987). Other common hospital-acquired infections caused by Klebsiella are urinary tract infections, surgical wound infections and infection of the blood. All of these infections can progress to shock and death if not treated early in an aggressive fashion (Malik et al., 2005).

Specific signs of an infected organ may pinpoint the primary or a metastatic site. Most neonates with early-onset GBS (and many with L. monocytogenes and Klebsiella) infection present with respiratory distress that is difficult to distinguish from hyaline membrane disease (Mondal et al., 1991). Periumbilical erythema, discharge, or bleeding without a hemorrhagic diathesis suggests omphalitis (infection prevents obliteration of the umbilical vessels). Coma, seizures, opisthotonos, or a bulging fontanelle suggests meningitis or brain abscess (Narla, 1985). Decreased spontaneous movement of an extremity and swelling, warmth, erythema, or tenderness over a joint indicates osteomyelitis or pyogenic arthritis. Unexplained abdominal distention may indicate peritonitis or necrotizing enterocolitis (particularly when accompanied by bloody diarrhea and fecal leukocytes). Cutaneous vesicles and mouth ulcers (Parrilio, 1991).

Klebsiella bacteria are generally resistant to many antibiotics, such as Penicillin. Often two or more antibiotics are used to eliminate the infection (Uppal, 1997). All Klebsiella strains causing primary liver abscess have a unique antimicrobial susceptibility pattern. They are resistant to Ampicillin, Ticarcillin, and Carbenicillin but susceptible to other antibiotics including all Cephalosporins and Aminoglycosides. Although multidrug-resistant strains of Klebsiella, whether community acquired or nosocomial, are not uncommon, these isolates had never been reported as the cause of primary liver abscess (Waddell et al., 1984). In early-onset sepsis, initial therapy should include Ampicillin, Penicillin G, Cefotaxime may be substituted for the Aminoglycoside. If foul-smelling amniotic fluid is present at birth, therapy for anaerobes Clindamycin, Metronidazole are added. Antibiotics may be changed as soon as an organism is identified (Amanda et al., 2009).


Blood specimens were collected during January to June 2009 from the hospitalized patients of The Children's Hospital and Institute of Child Health, Lahore with suspected sepsis. They required blood culture, sensitivity testing and were of both sexes in the age group of newborn to 7 months.

Following procedures were performed.

Collection of Samples

The prepared aerobic culture bottles were taken to the beside of the patient.The blood was drawn preferably from a peripheral vein, but if it was not possible then a femoral vein was chosen.The skin over the selected area was first cleaned thoroughly with 2% iodine,which was allowed to act for 1-2 minutes,and then the area was cleaned with 70% alcohol and allowed to dry.The blood was drawn in a sterile disposable plastic syringe.The sampling needle was discarded and the blood poured immediately in the culture bottles.1.5-2ml of blood was put in the culture bottle.After the addition of the blood the bottle is shaken gently to mix blood with the broth.The bottle was then labeled with the name of the patient and the date on date on which the sample was taken.Incubation of the blood sample was done at 37 oC.

Processing Of Samples

The broth was observed for the presence of turbidity, hemolysis and bubbles of gas and pellicle formation, after 18-24 hours of incubation at 37 oC.The following steps were undertaken on the first day after incubation.A loopful of culture broth,which was turbid,streaked on blood agar plate and MacConkey agar plate.The plates were incubated at 37oC for 24 hours.

On the second day,the plates were observed for bacterial growth.A separate individual colony representing the growth was studied for culturl characteristics.The characters noted were the size,shape and margins of the colony: Pigmentation and hemolysis was looked for around the colonies.

A smear of the colony was made,stained and examined under the microscope for staining reaction.

Sensitivity test was put up.

A sub-culture was made on blood agar plate from a separate colony to obtain a pure culture.Mix growth of more than one type is not considered in this study so that it indicates contamination.In case of no growth,the bottles were further incubated for 7 days and examined at the alternative days.On the third day,the following tests were done from the sub-culture plate.

Identification of Bacteria

For preliminary identification of bacteria the following tests were done:

Gram staining for morphology and staining reaction.

Catalase test


Gram's Staining

A Gram's stained smear of culture provides preliminary information about the type of organism present. Most bacteria can be differentiated by their Gram reaction due to differences in their cell wall structure. Gram positive organisms retain primary stain after decolourization and appear dark purple or blue. The Gram negative organisms are decolourized and appear pink or red as they take up the counter stain.

When certain bacteria are treated with one of the basic para-rosaniline dyes such as crystal violet and then with iodine they 'fix' the stain so that the subsequent treatment with a decolourizing agent such as acetone alcohol, does not remove the colour. Other organisms however are decolourized by this process. The species which retain dye are known as Gram positive whereas those which are completely decolourized are termed as Gram negative. In order to render the decolourized organisms visible and to distinguish them from those retaining colour, a contrast or counterstain is applied which is usually red so that violet Gram negative organisms can be easily differentiated.

A colony from the culture plate was emulsified in sterile distilled water and a thin preparation was made on a slide. The smear was dried and heat-fixed. Smear was covered with crystal violet stain for 30-60 seconds. Then stain was washed off with clean water. All the water was tipped off and the smear was covered with lugol's iodine for 30-60 seconds. Iodine was washed off with clean water. Smear was decolourized with acetone-alcohol (1:1) for few seconds. Then smear was washed with clean water. Smear was

covered with safranin stain for 2 minutes. The stain was washed off with clean water. Back of the slide was cleaned with the help of tissue paper and slide was placed in a draining rack for the smear to air-dry. Smear was examined microscopically first with the 40x object and then with oil immersion objective. Gram-positive bacteria appeared as dark purple while Gram negative bacteria appeared pale to dark red.

Catalase Test

This test demonstrates the presence of catalase, an enzyme that catalyzes the release of oxygen from hydrogen peroxide. Therefore, it is used to differentiate bacteria which produce catalase such as Staphylococci from non-catalase producing bacteria such as Streptococci.

One of the end products of aerobic carbohydrate metabolism is hydrogen peroxide, which if accumulates is lethal to bacteria. Catalase enzyme present in certain bacteria converts this hydrogen peroxide into oxygen and water by the following reaction:

2H2O2 Catalase 2H2O + O2 (Gas bubbles)

2-3 ml of hydrogen peroxide solution was taken in a test tube. With a sterile wooden stick growth of the test organisms was removed and immersed in hydrogen peroxide solution and reaction was observed.

Active bubbling immediately after immersing the stick in solution showed a positive test.

3. Oxidase Test

This test was carried out to differentiate Enterobacteriaceae which are all negative from species which give positive test such as Neisseria, Pseudomonas, Aeromonas and Alcaligenes etc.

A piece of filter paper was placed in a clean petri dish and 2 or 3 drops of freshly prepared solution were added. With a wooden stick a colony of the test organisms was removed and smeared on the filter paper. Development of blue- purple colour was noted. Blue-purple colour within 10 seconds showed a positive test (Cheesbrough, 2000).

4. API 20 E

API (Analytical Profile index) 20 E is a standardize identification system for Enterobacteriaceae and other non-fastidious, Gram negative rods which uses 21 miniaturized biochemical tests a database. The API 20 E strips consisted of 20 microtubes containing dehydrated substrates. Those tests were inoculated with a particular bacterial suspension that reconstituted the media. During incubation, metabolism produced the change in colour that was either spontaneous or revealed by the addition of reagents. The reactions were read according to the manufacturer's instructions.

Around 5.0 ml of distilled water was filled into the bottom of the tray to create a humid atmosphere. Three to four single isolated colonies were mixed into a 5.0 ml ampule of API NaCl 0.85%. The bacterial suspension was distributed into the tubes of the strip. All the tubes were half filled except the CIT, VP and GEL which were filled completely. The ADH, LDC, ODC, H2S, and URE tubes were further overlaid with mineral oil. Then it was covered with lid and was placed in an incubator at 36 ±2 oC for 18-24 hours. After that one drop of TDA, James, VP-1 and VP-2 reagent was added into the tubes containing TDA, IND and VP respectively. The reactions of the strip were read according to the reading table (Appendix-2). Each group of three tubes was valued by a number as 1, 2 and 4 respectively. A seven digit code is compared with the API 20 E identification book or the software.

Biochemical Tests

From the results of preliminary identification test, further biochemical tests were put up. The Gram negative rods were subjected to the following routine biochemical tests:

Urease production.

Citrate utilization test.

Triple sugar iron reaction.

Urease Test

Many species of bacteria possess an enzyme urease which can decompose urea to carbon dioxide and ammonia.

NH2- CO -NH2 + H2O 2NH3 + CO2

The presence of this enzyme can be detected by growing the organisms in the presence of urea and then testing for alkali production pH indicator. The ammonia produced reacts in solution to form ammonium carbonate; this increases the pH and the indicator phenol red changes colour form yellow to red.

The slants were inoculated with a pure culture of the organism, and incubated at 37oC for 24 hours. Positive result is indicated by change in colour of the medium to deep pink.

Citrate Utilization Test

This test is used for the differentiation of the family Enterobacteriaceae based on whether citrate is utilized or not as the sole source of carbon. Simmons citrate agar was used. A well isolated colony was used to inoculate lightly the slant of the citrate agar and the butt of the medium was inoculated by stabbing. The tube was incubated at 37oC for 48 hours.

Some bacteria utilize citrate as sole source of carbon. The medium used to detect citrate utilization was devoid of protein and carbohydrates as sources of carbon. Utilization of citrate by test bacteria was detected by production of alkaline byproducts which change the colour of medium from green to bright blue. A positive test is represented by the development of a deep blue colour. The test is also read as positive when there is visible colonial growth along the inoculation streak line without the development of blue colour.

Triple Sugar Iron (TSI)

TSI is a composite medium for the differentiation of Enterobacteriaceae according to their ability to ferment lactose, sucrose, glucose, and to produce hydrogen sulphide.

A well isolated colony on CLED agar was touched with the end of a straight inoculating wire. It was stabbed into the butt of the medium, extending to within 4 mm of the bottom. While removing the wire from butt, the slant was also streaked. The tubes were incubated at 37oC for 24 hours.

Carbohydrate fermentation occurs with acid production with or without production of gas. The phenol red indicator in the medium changes colour of medium to yellow at an acidic pH and red at alkaline pH. When carbohydrates are fermented, the acid produces a fall in pH and medium becomes yellow. The organisms which do not ferment carbohydrates produce alkalinization due to release of amines by degradation of aminoacids. Therefore, the pH increases and the medium become red.

Ferrous sulphate in the medium is used as an indicator for colourless hydrogen sulphide gas. Sulpher items are supplied from sodium thiosulphate. The hydrogen sulphide reacts with ferrous sulphate in an acidic environment to produce insoluble black precipitate, ferrous sulphide, this causes blackening of the medium.

Antimicrobial Sensitivity Test

All organisms isolated were tested against various antibiotics in vitro by Kirby-Bauer disc diffusion method. Colonies were picked up by a sterilized wire loop and put on the centre of the Muller Hinton agar plate. Then with a sterile cotton swab the colonies were streaked on the plate starting from the centre. The appropriate antimicrobial discs (Augmentin, Cephalexin, Cefixime, Ciprofloxacin, Ceftazidime, Naladixic Acid, Nitrofurantoin, Pipemidic Acid, Cefizox, Imipenem, Oxacillin and Vancomycin) were placed evenly distributed and the plate incubated at 37oC overnight. After overnight incubation the diameter of each zone of inhibition was measured in mm. The endpoint of inhibition was where the growth starts. Interpretation of zone sizes of each antimicrobial disc was made, as Sensitive or Resistant using interpretation chart of zone sizes.

The present study was carried out in the Microbiology department of The Children Hospital and Institute of Child Health, Lahore. Blood samples were analyzed to determine the frequency and sensitivity pattern of Klebsiella during the six month time period starting from January to June 2009. Among the 12000 samples studied, 1070 samples showed positive results. The frequency of gram positive was found to be 27.6% and gram negative was 72.3% (Table-4.1). The results of present study are in accordance with the one carried out by Rizwan et al., (2004) at the Neonatal Unit of Ghurki Trust Teaching Hospital Lahore, where Gram negative organisms were isolated from 80% of the cases. E. coli was the commonest isolate, followed by Klebsiella (76%) and Pseudomonas (13%).

In current study, among the 1070 positive samples, the frequency of Klebsiella was 408 (38.31%) (Table 4.2). Similar findings were attained by Bhutta, (1996), where the E. coli, Klebsiella and Pseudomonas constituted 77% of the total organism isolated. According to this analysis, more than 60% of the cases of sepsis are due to gram negative isolates and the most commonly isolated organism is Klebsiella (41%).

Haroon et al., (2007) reported the frequency, causative organisms and susceptibility pattern of nosocomial bloodstream infections in children at paediatric intensive care unit of Children's Hospital Lahore, from January to December 2004. Majority of isolates (77%) were gram-negative bacteria; Klebsiella being the most common isolate (36%). The results of this study are in accordance with the present study..

Shah, (2002) Studied the prevalence of enterobacteriacae in Quaid-i-Azam University, Islamabad and reported the Escherichia coli (35%) as most common organism followed by Klebsiella pneumoniae (25%), Enterobacter cloacae (15%), Proteus mirabilis (10%), Proteus vulgaris (5%) and Citrobacter freundii (10%). These results are are not compatible with my results. Similarly the research work by Zulfiqar et al., (1989) at

 the Aga Khan University Hospital, Karachi reported that the most common organisms causing sepsis were Klebsiella species (53%) and Escherichia coli (10%), whereas other organisms included Salmonella parathypi (21%), Group A Streptococcus (21%). The frequency in this hospital was much higher than the current study.

A very low sensitivity and prevalence of klebsiella spp. were noticed by Rehman et al., (2003) in the Department of Pediatrics, Khyber Teaching Hospital, Peshawar. Escherichia coli was the most common organism found (36.6%), followed by Staphylococcus aureus (29.5%), Pseudomonas (22.4%), Klebsiella (7.6%), and Proteus (3.8%).

In the present study  antimicrobial sensitivity pattern for Klebsiella revealed highest sensitivity to Pipracillin-Tazobactum (87.74%) and Ciprofloxacin (81.37%) and Amoxicillin + Calvulanic acid (62.25%) (Table-4.3). Similar antibiotic sensitivity pattern of neonatal sepsis was demonstrated by Malakan and Momtazmanesh at Kashan University of Medical Sciences, Iran. They reported sensitivity of the isolates to Cefotaxime. They reported the percentage of extended-spectrum beta-lactamase (ESBL)-producing Klebsiella species that were resistant to all the third generation Cephalosporins, was 23%.

High resistivity was observed with Cefuroxime (4.41%), Cefixime (6.61%), Gentamycin (10.53) and Cefetazidime (16.17%) (Table-4.3). Similar results by Jain et al., 2007 reviewed blood samples of 728 neonates with suspected sepsis and reported that 36.6% of Klebsiella species were resistant to these antibiotics.

Thaver et al., (2009) reported that among Klebsiella species, almost all were resistant to Ampicillin, 45% to Cotrimoxazole, and 66% to third generation Cephalosporins. Resistance to Gentamicin was much higher among Klebsiella species (60%). These results are in accordance to my study.

The case records of all neonates admitted to the neonatal unit of Al Wasl Hospital (Dubai) in a period of 60 months were analysed by Ayman et al., (1995). One hundred and six neonates had confirmed sepsis. Klebsiella pneumoniae was found to be the causative organism in 24% of cases. The frequency of Klebsiella was much lower in these countries than in Pakistan. The reason behind is that the spectrum of organisms that cause neonatal sepsis changes over times and varies from place to place (Malakan and Momtazmanesh, 2004). Gram negative organisms had been the most common cause at present, in the developing countries, because these organisms have developed multi-drug resistance over the last two decades (Gladstone et al,. 1990). The reasons for this resistance is the indiscriminate use of antibiotics, over the counter sale of antibiotics and ineffective infection control in maternity centers (Samaie, 1997).

Similarly Rizwan et al., (2004) at the Neonatal Unit of Ghurki Trust Teaching Hospital Lahore, demonstrated that Klebsiella showed a low sensitivity to Cefuroxime, Gentamicin, and Ceftrixone, while good sensitivity to Amikacin, Ceftazidime and Ciprofloxacin. These results are exactly similar to our own study carried out at Children's hospital.

Doctors, staff and other healthcare workers in hospitals can do much to reduce nosocomial infections through identification and control of predisposing factors, education and training of hospital personnel, and limited microbial surveillance (Jalil et al., 1997).