INTRODUCTION

 

Nosocomial infections (NI) are acquired after hospital admission by more than 24 hours, or within 48 hours of discharge.¹ The incidence in United States is more than 200,000 patients per year. Measuring the risk for NI could also be the first step in setting benchmarks of anticipated performance to evade NI. Resistance to antimicrobial is considered as a key reason of high morbidity, mortality and health care cost.^1,2

As well, antibiotics overuse, poor implementation of infection control measures, prolonged hospitalization, prior admission to intensive care units and invasive procedures are extra risk factors. Also, diagnosis on admission and severity of illness has been reported in high mortality. ^3,4  The current study was designed to detect the frequency of NI according CDC, common microorganisms, risk factors and mortality rate.

PATIENTS AND METHODS

 

Study population: This prospective study was conducted in the neurology department (which contains 7- bed ICU and 100-bed wards) at Mansoura University Hospital. All enrolled patients in the current study had symptoms that met the criteria of NI but patients who died or whose length of stay was less than 24 hours were excluded.

The clinical specimens from patients who developed unexplained fever >38ºC, leukocytosis, new infiltrates on chest X-ray, tracheal secretions, turbid urine or bed sore infection were submitted for bacterial cultures at microbiology diagnostic and infection control unit. The prospective external or internal mortality risk factors were documented and the severity of illness was assessed using GCS.

 

Samples collections:

Blood stream infections (BSI): Fresh venipuncture blood (5–10 ml) was inoculated into 50 ml blood culture bottles (Micrognost, Biotech). A blind subculture was done 18 hours later; if no growth, the bottles were examined daily for 7 days.

Urinary tract infections (UTI): Wet mount urine preparation was examined microscopically to detect significant pyuria, yeast cells or bacteria. A Gram stained smear was also used for bacteria. Urine and tips of indwelling urinary catheters were cultured simultaneously.

Respiratory tract infections (RTI): In cases with suspected hospital acquired pneumonia, empyema and ventilator associated pneumonia (VAP), the sputum, respiratory secretions, tips of endotracheal tubes (ETT) and swabs from suction catheter were cultured on same media.

Bed sore infections: In bed sore infections, the edges of the bed sore were compressed to ooze then swabbing was done using sterile cotton swabs and cultured on cultures media.

Central nervous system infections: The cerebrospinal fluid (CSF) was collected aseptically by lumbar puncture from suspected cases and after centrifugation; the sediment was inoculated on same cultures media.

Identification: The organisms were identified according to the colony characteristics and biochemical reactions. API 20 E and API 20NE system (bioMerieux, Marcy l'Etoile, France) were used to identify Gram negative isolates, while culture characteristics, catalase and coagulase activity were employed to confirm Gram-positive organisms identity. Simultaneously, a Gram stained smear was also prepared. In addition, germ tube test was done to differentiate Candida albicans from non albicans Candida.  

Antibiotic susceptibility testing: Antibiotic susceptibility to Pencillin, Ampicillin, Sulbactam-ampicillin, Piperacillin, Aztreonam, Meropenem, Vancomycin, Cefradin, Cefuroxime, Cefotaxime, Ceftazidime, Ciprofloxacin, Norfloxacin, Trimethoprim-sulfamethoxazole, Gentamicin, and Amikacin was determined using Kirby-Bauer method and the results were interpreted as per National Committee for Clinical Laboratory Standards (NCCLS) guidelines.⁵

Investigation of detected Enterbacter cloacae outbreak: Swabs were taken at the time of E. cloacae outbreak from frequently touched items or surfaces with suspected transmission and from hands of health care workers dealing with the outbreak cases then cultured on Mac-Conkeys plate and incubated overnight at 37°C.

DNA of E. cloacae strains was isolated according to the method described by Lema et al. Subsequently, inter-repeat PCR was performed using the Enterobacterial Repetitive Intergenic Consensus (ERIC) 2-primer (sequence: 5 AAGTAA GTG ACT GGG GTG AGCG 3). PCR products were electrophoresed according to standard protocols.^6,7

Statistical Analysis

The demographic, clinical, and technical data were collected using a ‘data collection form’ and entered into a computerized database before statistical analysis. Continuous variables were compared using analysis of variance for repeated measures. P-value less than (0.05) was considered statistically significant. All data were expressed as mean ± standard deviation (SD) or patient’s number (n) and percentage (%) as appropriate. Significant predictors on bivariate analysis were entered into a logistic regression model using forward Wald method to predict the independent predictors of infection. Relative risk (RR) and confidence interval (95% CI) were calculated.

 

RESULTS

 

The present study enrolled 1237 samples from 792 patients (412males and 380 females with mean age 49.6±18.2 years). In addition to potential risk factors (Table 1), the most common neurological diseases on admission were stroke, epilepsy, demyelinating diseases, myasthenia gravis or GBS under plasmapharesis, coma and neuropathy (Table 2).

NI was confirmed in 162 out of 792 patients (20.5%). The mean age was 51.39±14.9 years with age ranged from 11 to 89 yo. Male to female ratio was 1.2:1. Average duration of stay in the neurology department was 3.8 days per patient (3-18 days).

Among these specimens, 244 specimens yielded different nosocomial pathogens; the remaining isolates failed to grow or were contaminated on arrival. The samples were distributed as follow: 93 blood, 61 urine, 17 bed sore, 67 respiratory, and 6 CSF samples from patients with suspected BSI, UTI, SSI, LRTI and CNS infections respectively.

The risks of developing Nosocomial UTI were detected among forty-nine patients with catheter in-situ for more than seven days developed as compared to twelve with catheter in-situ for less than seven days (‘p’ <0.05).The risk was also a higher among patients with more than three days of ICU stay than ward (‘p’ <0.05). 120 out of 792 patients (15.2%), 37 out of 162 NI patients (22.8%) and 83 out of 630 patients without NI as a control of non infected patients (13.2%) died that denoted the mortality rate was statistically significantly high among patients with nosocomial infections (Table 3).

The organisms that caused NI were Klebsiella pneumonie (K. pneumoniae) (48) 26.8%, Escherichia coli (E. coli) (31) 17.3%, Pseudomonas aeruginosa (P.aeurginosa)(24) 13.4%, Candida albicans (22) 12.3%, Enterobacter cloacae (17) 9.5%, Proteus spp. (16) 8.8%, methicillin resistant Staphylococcus aureus (MRSA) (11) 6.1%, Staphylococcus aureus (6) 3.4%, Coagulase negative Staphylococci (4) 2.2%. Single organism infection happened in 147 (90.7%) cases and multiple bacterial species in 15 cases (9.3 %). Eight patients (4.9%) had the same isolates (2 MRSA, 2 Pseudomonas, 2 Enterobacter, 1 E. coli and 1 Klebsiella) and parallel isolated from blood and urine. Four cases of them (50%) died.

Most of the isolates were resistant to the majority of the used antibiotics especially Gram negative bacilli and MRSA strains but no bacterial growth from CSF samples was detected.

In multivariate analysis, nosocomial infection (P<0.001), mechanical ventilation (P= 0. 003), presence of two or more organ diseases (P<0.05), Ischemic/hemorrhagic stroke, Status epilepticus (P<0.05), Multiple risk factors (P<0.001), GCS score (3-8) (P=0.01) and old age (>65 yo) (P<0.05) were identified as risk factors for mortality (Table 3). Logistic regression model for risk factors showed high mortality among patients had low Glasgow Coma Scale (3-8) (p=0.01), mechanical ventilation (p=0.003), nosocomial infection (p=0.001) and multiple risk factors (p=0.0001) with CI 95% (Table 3).

For a period of one week, 3 patients developed Enterbacter cloacae UTI. The isolates revealed resistance to pencillins, aminoglycosides and second and third generation cephalosporins, while all showed 100% sensitivity to meropenem. Enterbacter cloacae were typed using ERIC-PCR (Figure 1) and showed the same genotype that yielded in the 3-isolates and hands of one nurse.⁶

 

 

Table 1. The common risk factors among studied patients.

 

Risk Factor	n=792	100%
DM	245	30.9
Cardiac diseases	201	25.4
Hypertension	341	43.1
Smoking	182	22.9
Prior CVS	102	12.9
Hyperlipidemia	79	9.9
Impaired LFT	166	20.9
Impaired RFT	87	10.9
High ESR	277	34.9
Anemia	142	17.9
No obvious RF	261	32.9

 

 

Table 2. Types of diseases on admission.

 

Diagnosis on admission	N	%
Cerebral hemorrhage	53	6.7
Cerebral infarct	251	31.7
Subarachnoid hemorrhage	12	1.5
Myasthenia gravis	7	0.9
Epilepsy	127	16
Confusional state	61	7.7
PNS disorders	103	13
OND	186	23.5
Total	792	100

CVS= cerebrovascular stroke, LFT= liver function tests, RFT= renal functions tests, RF= risk factor

Table 3. Variables associated with high mortality among NI patients.

 

Variable	P-value	RR	Z statistic	95%CI
Nosocomial infection	0.0001	1.76	4.52	1.38–2.24
Mechanical ventilation	0.003	0.62	2.93	0.45-0.85
Presence of two or more organ diseases	0.0001	3.48	11.16	2.79-4.33
Glasgow Coma Scale score (3-8)	0.01	0.67	2.48	0.49-0.93
Ischemic/hemorrhagic stroke	0.0001	3.43	10.91	2.75-4.27
Status epilepticus	0.03	0.73	2.13	0.53-0.97
Multiple risk factors	0.001	3.57	11.41	2.87-4.43

 

 

 

 

 

Figure 1. Representative ERIC-PCR amplification. Lane 1 is the 10000bp DNA ladder (200, 400, 600, 800, 1000, 1500, 2000, 2500, 3000, 4000, 5000, 6000, 8000 and 10000 bp) Lane 3, 4 and 6 represent E. cloacae outbreak isolates. Lane 8 shows E. cloacae from the hands of the source.

 

 

DISCUSSION

 

The patterns of microbial infection by the endogenous or exogenous routes and antibiotic resistance differ extensively from one hospital or country to another. This is a very multifaceted issue, encompassing hygiene, cross-transmission, devices, antibiotic therapy, host immunity and emergence of antibiotic resistance.⁸

In the current study, we investigated the frequency of NI among patients in neurology department. NI has greater impact on morbidity with its direct health cost consequences and mortality. NI was seen in 162 out of 792 patients (20.5%) while in certain centers in Germany showed 4.2% and 8.7%, while Iranian study revealed 43.2%. The relative high incidence in present study may be owed to the insufficient level of care.^8,9,10

As an evidence of importance of such studies, we find that K. pneumonia was the most frequent organism discovered in our neurology department positive samples while coagulase negative staphylococci was the least frequent. In a recent study, similar department but different country (in Turkey), the coagulase negative Staphylococci was the most prevalent organism and Klebsiella was the least.⁴ Its source was blood secondary to pneumonia infection rather than UTI. In the present study, population was mainly stroke patients (40%) where pneumonia is one of its common complications.^5,9

Hilker et al. in their study found a 21% incidence of NI pneumonia in acute stroke patients treated in neurology ICU. One of the most important factors putting patients at an increased risk of pneumonia is a decreased level of consciousness that leads to an attenuation of protective reflexes, impaired functioning of lower esophageal sphincter, delayed gastric emptying as well as worsening of the coordination between swallowing and breathing.^11,12

In the current study; old age, female gender, disease type (stroke), immuno-compromised status and co-morbidity showed augmentation of the risk of acquiring NI. These results are concurred with the results of other published data which documented a higher risk for adult female patients in developing bacteriuria, the elderly and critically ill patients in developing blood stream bacteremia or septicemia. Patients hospitalized due to stroke are usually elderly patients and they have additional risk factors such as immobility, debilitating and intensive use of steroids making them a high risk group of developing nosocomial infection.¹³

Moreover, in present study, NI-positive patients, those with coma (GCS = 3-8) were at higher risk of mortality. Candida NI was reported in 12.3% of NI-positive  patients, a finding that is not only high but also imply the severity of illness in such patients as patients who are critically ill and in medical and surgical ICUs have been the prime targets for opportunistic nosocomial fungal infections primarily due to Candida species.¹⁴ As regard antibiotic susceptibility, meropenem appeared to be the most effective antibiotic against Klebsiella, E. coli and Enterbacter whereas ciprofloxacin was more potent against Proteus and Pseudomonas. While all isolated Staphylococci were susceptible to vancomycin.¹⁰

Aforementioned data are so beneficial for tailoring empirical antibiotic therapy. Since NI resulted from a diversity of Gram-negative as well Gram-positive bacteria, which may sway from time to time and from one institution to the other. No antibacterial agent would be appropriate for all clinical location; therefore each institution must have its own recent available information.⁸ This was nicely demonstrated in current study when eight patients (4.9%) had the same isolated organisms (2MRSA, 2 Pseudomonas, 2 Enterobacter, 1 E. coli, and 1 Klebsiella) from the culture of urine that were collected through the catheter and the blood.^8,9 Obviously urosepsis, carried high mortality rate (50%).^15,16

According to the collected data, it was deduced that the most significant risk of acquiring NI was the number of hospital stay days. Prolonged hospitalization more than 3 days especially in ICU had higher risk of acquiring NI (‘p’ value =0.01). The extended hospital stay increases the colonization of skin and environment of the patient that may be responsible for higher NI incidence.^9-18

NI-mortality rate could be referred to the underlying critical patients’ conditions but there is statistical significance differences were detected between those with NI and those without. The current study mortality rate showed 120 patients died out of 792 enrolled patients (15.2%), while 37 patients out of 162 with NI (22.8%) and 83 patients out of 630 without NI (13.2%) died. The mortality was statistically significantly higher among NI patients especially patients who had history of coma (GCS 3-8), mechanical ventilation, and multiple risk factors or stayed in ICU for long time.^4,17,18

The current result is much comparable to the result of Arunodaya GR who recorded 29.5% mortality rate. While, mortality rate in Neurological ICU in Turkey was 62.5%.¹⁷ 

There was outbreak of Enterobacter cloacae. It was emerged as 3-isolates with the same antibiotic susceptibility pattern. Molecular typing confirmed the entire isolates of the same clone. Surveillance cultures incriminated the hands of a nurse as a source of infection. These results confirmed that many single-clone outbreaks could be resulted from cross-transmission via healthcare workers (HCWs).¹⁹

Furthermore over-crowding and understaffing with increased workload in the period of E. cloacae outbreak have also been identified as auxiliary causes. These issues raised the infection rate because of slippage of healthcare worker, aseptic technique or inadequate hand washing which gave us an idea about hygienic means that away from being optimal.^19,20 

The antibiotic resistance profile of the outbreak isolates in current study was resistance to aminoglycosides, all the β-lactam antibiotics tested with the exception of carbapenem. The isolates were uniformly sensitive to carbapenem as well as the quinolone.¹⁷ It was verified that fortification of hygiene practices especially hand hygiene, the use of proper disinfection, antibiotic restrictions and limitation of new admissions, led to outbreak control.^17-20

In conclusion, NIs were common among patients with neurology disorders. K. pneumonia and E. coli were the commonest organisms in present study. Furthermore, prolonged hospital stay, external devices and underlying diseases were the most important risk factors that increase the incidence of NI while high mortality rate were more noticed with coma, mechanical ventilated patients, those with multiple risk factors and nosocomial infections. Also, the best empiric antibiotic therapy for NI among neurological disorders patients were imipenem and cefepime.^{9, 17}

It seems reasonable that application of rigorous infection control measures considering cost effectiveness had a great impact on reduction of nosocomial infection, mortality rate, limitation of the emergence of antibiotic resistant organisms, reduction of hospital stay and rapid management of detected outbreak.

 

REFERENCES

 

1.      Garner JS, Jarvis WR, Emori TG, Horan TC, Hughes JM. CDC definitions for nosocomial infections. Am J Infect Control. 1988;16:128-40.

2.      Minhas P, Perl TM, Carroll KC, Shepard JW, Shangraw KA, Fellerman D, et al. Risk factors for positive admission surveillance cultures for methicillin-resistant Staphylococcus aureus and vancomycin-resistant enterococci in a neurocritical care unit. Crit Care Med. 2011 ;39(10):2322-9.

3.      Wibbenmeyer L, Danks R, Faucher L. Prospective analysis of nosocomial infection rates, antibiotic use, and patterns of resistance in a burn population. J Burn Care Res.2006;27(2): 152-60.

4.      Cevik MA, Yilmaz GR, Erdinc FS, Ucler S, Tulek NE. Relationship between nosocomial infection and mortality in a neurology intensive care unit in Turkey. J Hosp Infection .2005; 59: 324-30.

5.      National Committee for Clinical Laboratory Standards: Performance standards for antimicrobial disk susceptibility testing, 8th ed. M2-A8. NCCLS, Wayne, PA.; 2003.

6.      Stephen M P, John Bryan, Michael F C, BDS. Enterobacterial repetitive intergenic consensus polymerase chain reaction typing of isolates of Enterobacter cloacae from an outbreak of infection in a neonatal intensive care unit. Am J Infect Control. 2000;28:123-9.

7.      Lema MW, Arnold Brown A, Calkins JH. A general method for the extraction of DNA from bacteria. Journal of Microbiological Methods. 1994; 19: 167-72.

8.      Wagenlehner FME, Naber K G. Emergence of antibiotic resistance and prudent use of antibiotic therapy in nosocomially acquired urinary tract infections. International Journal of Antimicrobial Agents. 2004; 23: 24-9.

9.      Dettenkofer M, Ebner W, Els T, Babikir R, Lucking C, Pelz K, et al. Surveillance of nosocomial infections in a neurology intensive care unit. J Neurol. 2001; 248: 959-64.

10.    Taher MT, Golestanpour A. Symptomatic nosocomial urinary tract infection in ICU patients: identification of antimicrobial resistance pattern. Ir J Clini Infect Dis. 2009; 4 :25-9.

11.    Hilker R, Poetter C, Findeisen N, Sobesky J, Neveling M, Heiss WD. Nosocomial pneumonia after acute stroke: implications for neurological intensive care medicine. Stroke. 2003; 34: 957-981.

12.    Katzan IL, Cebul RD, Husak SH, Dawson NV, Baker DW. The effect of pneumonia on mortality among patients hospitalized for acute stroke. Neurology. 2003; 60: 620-5.

13.    Tekin R, Dal T, Cevik M, Bozkurt F, Deveci O, Tekin A. Fourteen Year Surviellance of Nosocomial Infections in neurology unit. J Neurol. Res.2012,2(3), 88-92

14.    Pappas PG, Rex JH, Lee J, Hamill RJ, Larsen RA, Powderly W, Kauffman CA, Hyslop N, Mangino JE, Chapman S, Horowitz HW, Edwards JE, Dismukes WE. A prospective observational study of candidemia: epidemiology, therapy, and influences on mortality in hospitalized adult and pediatric patients. Clin Infect Dis. 2003; 37 (5): 634-43.

15.    Ramirez Barba EJ, Rosenthal VD, Higuera F, Oropeza MS, Hernández HT, López MS, et al. Device-associated nosocomial infection rates in intensive care units in four Mexican public hospitals. AJIC: Am J Infect Control. 2006:  34 (4): 244-7.

16.    Poisson SN, Johnston SC, Josephson SA. Urinary tract infections complicating stroke: mechanisms, consequences, and possible solutions. Stroke. 2010;41(4):e180-4.

17.    Arunodaya GR. Infections in neurology and neurosurgery intensive care units. Neurol India. 2001 49; Suppl 1:S51-9.

18.    Cevik MA, Yilmaz GR, Erdinc FS, Ucler S, Tulek NE. Relationship between nosocomial infection and mortality in a neurology intensive care unit in Turkey. J Hosp Infection. 2005; 59: 324-30.

19.    Grattard F, Pozzetto B, Berthelot P, Rayet I, Ros A, Lauras B, et al. Arbitrarily primed PCR, ribotyping, and plasmid pattern analysis applied to investigation of a nosocomial outbreak due to Enterobacter cloacae in a neonatal intensive care unit. J Clin Microbiol. 1994;32: 596-602.

20.    Manzur A, Tubau F, Pujol M, Calatayud L, Peña C, Sora M, et al. Nosocomial outbreak due to extended-spectrum-beta-lactamase- producing Enterobacter cloacae in a cardiothoracic intensive care unit. J Clin Microbiol. 2007; 45(8):2365-9.