Point prevalence survey to estimate antimicrobial use in a tertiary care university hospital in Pakistan using WHO methodology: findings and implications

To the Editor,

Recently in this journal, considerable variation in antibiotic prescribing and antibiotic resistance among elderly citizens in two adjacent Nordic regions was reported [1]. The authors discussed differences in prescriber preferences and guidelines and suggested more direct attention towards antimicrobial stewardship to help inform and motivate prescribing behaviours. Based on data of well-known high consumption of broad-spectrum antibiotic in hospital settings of Pakistan, we conducted a point prevalence survey in different wards. The information is supposed to become a basis for antimicrobial stewardship programs.

This cross-sectional observational study was conducted in Mayo Hospital Lahore during 19th October 2020 to 29th October 2020. A validated World Health Organization point prevalence survey method was used to collect the data. Simple random probability sampling technique was used to select the patients. The study setting for this PPS study in Punjab, Pakistan was Mayo Hospital which is leading tertiary care teaching hospital of the most populous city Lahore affiliated with King Edward Medical University which is the prestigious and historical institution of Indo Pak subcontinent where excellence in medical education is pursued and nurtured since its inception in 1871. A validated World Health Organization point prevalence survey method was used to collect the data. All hospitalized patients admitted before or at 8:00 am with no plan of discharge from the ward at the time of survey were included irrespective of whether they were receiving antibiotics or not. Emergency cases, acute care wards, day care surgery, dialysis units and chemotherapy units were excluded from the survey.

The data regarding patient demography age, sex, admission detail, laboratory and radiological investigations, the use of invasive devices, documentation of antimicrobials including antibiotic type, dose, indications, route of administration, duration, their empirical/targeted use with the evidence of culture and sensitivity reports, stop/review order of antimicrobial after 48 h, availability and compliance of antimicrobial guideline policy were collected on online survey form using KOBOCOLLECT software. The access to KOBOCOLLECT software was given by WHO and also training given to survey team before starting survey. Data were collected from patient’s files in all eligible wards and also interviewing from attending doctors and nurses regarding patients missing information. Each ward was completely surveyed within 1 day. At the end of point prevalence survey, the collected data on KOBOCOLLECT Performa were analyzed.

A point prevalence survey of 671 patient was done. Out of 671 patients, 247 files were from adult surgical ward (36.81%), 184 adult medical ward (27.42%), 58 adult high risk ward (8.64%), 38 paediatric medical ward (5.66%), 34 neonatal medical ward (5.07%), 25 paediatric surgical ward (3.73%), 24 paediatric high risk ward (3.58%), 23 paediatric intensive care unit (3.43%), 22 neonatal ICU (3.28%) and 16 patients in adult ICU (2.38%) (Table 1). Among them, 403 (60.06%) were males and 268 (39.9%) were females. Unexpectedly, 551 (89.97%) patients were on antibiotics while 121 patients (18.03%) were not on antibiotics. All 551 patients were being prescribed 1054 different antibiotics. Out of these, 1019 (96.67%) were being given through parenteral route while 35 (3.32%) were given through oral route. It was noted that mostly multiple antibiotics were used on single patient with an average 2 antibiotics per patient. According to types of antibiotics, the most commonly prescribed antibiotic was ceftriaxone (47.91%, followed by metronidazole (23.04%), amikacin (21.96%), amoxicillin/amoxicillin plus clavulanic acid (17.42%) and moxifloxacin (13.79%). Different indications to use antibiotics were surgical prophylaxis (39.67%), medical prophylaxis (27.84%), community acquired infections (23.29%) and hospital acquired infections (4.45%). Out of total 419 antibiotics used as surgical prophylaxis, 227(54.17%) were prescribed as more than one dose for more than 24 h. A total of 953 (90.41%) antibiotics were used empirically while only 101 (9.5%) antibiotics were used as targeted therapy. Out of 551 patients who were on antibiotics, specimen of only 104 patients were sent for culture sensitivity as per documentation. Out of these 104 specimens, reports of 51 patients retrieved while 51 reports were pending while other reports were missing. Specimens sent for culture include 48 blood, 29 urine, 30 wound, 7 respiratory and 9 others samples for culture. Out of 51 reports, there were no growth in 31 (60%) samples, while positive growth in 20 (40%) samples. The positive growth includes Staphylococcus aureus (35%), Escherichia coli (25%), klebsiella (15%), while pseudomonas, candida, salmonella, proteus and proteus mirabilis were present in one sample each. In 500 (74.52%) patients, there were no stop/review orders and the verbal stop/review orders was only in 36 (5.37%) patients. The details of results are presented in Tables 1 and 2.

Table 1. Wards level data.

Types of wards	No. of patients	Percentage
Adult surgical wards	247	36.81
Adult medical wards	184	27.42
Adult high risk wards	58	8.64
Paediatric medical wards	38	5.66
Neonatal medical wards	34	5.07
Paediatric surgical wards	25	3.73
Paediatric high risk wards	24	3.58
Paediatric ICU	23	3.43
Neonatal ICU	22	3.28
Adult ICU	16	2.38

Table 2. Patients level data.

Parameters	Frequency	Percentage
Sex
Male	404	60.12
Female	267	39.79
Patients receiving antibiotics at time of survey
Yes	551	81.99
No	121	18.01
Route of administration
Parenteral	536	79.88
Oral	15	2.23
Types of antibiotics
Ceftriaxone	264	47.91%
Metronidazole	127	23.04
Amikacin	121	21.96
Amoxicillin clavulanic acid	96	17.42
Moxifloxacin	76	10.71
Tazobactam piperacillin	72	13.06
Meropenem	59
Vancomycin	39
Sulbactam cefoperazone	38
Cefotaxime	21	3.8
Ciprofloxacin	17	3.08
Indications
SP	419	33.48
MP	277	23.07
CAI	253	18.75
HAI	47	3.42
Others	58	2.98
No. of doses of antibiotics used for Surgical prophylaxis
>1 dose in more than 24 h	227	54.17
>1 dose within 24 h	62	14.79
1 dose only	60	14.31
No antimicrobial taken	60	14.31
No dose	10	2.38
Empirical/targeted
Empirical	953	90.41
Targeted	101	9.5
Stop/review order
No stop/review order	501	74.55
Yes, stop/review not documented	36	5.36
Yes, a stop/review order documented	14	2.08
Culture report
Present	51	9.25
Pending	51	9.25
Missing	2	0.36
Type of microbes
Staphylococcus aureus	7	35
E. coli	5	25
Klebsiella sp.	3	15
Pseudomonas sp.	1	5
Candida sp.	1	5
Others	3	15

Our results showed high use of antibiotics without rationale and antibiotic guideline policy of Mayo hospital. There was only one document for emergency use of antibiotics in emergency but overall, there was no documented policy of antibiotics applicable to all hospitalized patients. Only few wards had antibiograms done in their department and were using antibiotics according to them but no individual documented antibiotics guidelines were present in departments. Infection prevention and control policies were either not present nor they were adapted properly. Poor Infection prevention and control practices because of economic issue in the hospital lead to misuse and overuse of broad spectrum antibiotics [2]. Antimicrobial resistance is also increasing due to increasing trend of hospital acquired infections, overcrowding of patients in hospital and more urbanization [2,3].

In our study, 81.99% patients were on antibiotics at the time of survey while another study from Pakistan reported 77% patients receiving antibiotics at time of survey. This is much higher prevalence rates compared to studies reported from America, Turkey, chin and European countries [4]. Nevertheless, this prevalence rate is similar to the hospitals of some African and Asian countries [4]. In our study, 536 (79.88%) were on parenteral antibiotics while in another study also majority of patients were on parenteral antibiotics [5]. There is need to implement good ASP to decrease use of parenteral antibiotics [6]. In our study, majority of antibiotics were used empirically (90.41%) and only (9.5%) were used as targeted antibiotics according to culture and sensitivity reports. The reason for this was that specimen were not taken for culture and sensitivity before start of antibiotics despite availability of microbiological lab or if taken during use of antibiotics most of the results of no growth (in 31 cultures) of bacteria came. Also, another reason was that proper protocol for taking specimen for culture was not followed. In another study carried out in low and middle income courtiers 65% of reserve antibiotics were used empirically [7]. In a study done at tertiary care private hospital, 92% antimicrobials were used empirically comparable to our study [8]. In a study done in Thailand, 75.4% patients received antibiotics empirically [9]. Also, empirical use of antibiotics has been reported in other LMICs due to logistical challenges [5,10]. In our study, most commonly prescribed antibiotics was ceftriaxone (47.91%) followed by metronidazole (23.04%). In other published studies, most common prescribed antibiotics were also ceftriaxone (25.1%) and metronidazole(24.6%) which is comparable to our result [4]. In our survey, the most common indication for antibiotics was surgical prophylaxis (33.48%) followed by medical prophylaxis (23.07%) and community acquired infection (18.75%). In another study community acquired infection and surgical prophylaxis were also main indications for antibiotic use [4]. The reason for more use of antibiotics as surgical prophylaxis was that majority of patients have no evidence of culture proved infection so their antibiotics were considered as surgical prophylaxis. In our study, antibiotics duration for surgical prophylaxis was more than one day. We also noticed in our study that majority of antibiotics were not given at time of induction. As we have noticed that single doses surgical prophylaxis was very less in number and also more than one day surgical prophylaxis is high in percentage despite single-dose recommendations for most indications. So, there is need to apply antimicrobial stewardship to decrease unnecessary antibiotic use for surgical prophylaxis in order to reduce antimicrobial resistance. According to another study, postoperative infections were not decreased due to use of antibiotics more than 2 days as compared to less than 2 days. In our study, indications for antibiotics were documented in only 11.9% while in another study documentation of antibiotics was 97% [4]. This may be due to overcrowdings of patients, lack of proper structured files for documentation and traditional attitude.

We observed high use of antibiotics both for prophylactic and therapeutic purposes for hospital acquired infections and in majority of cases antibiotics were used empirically without any written documentation of stop or review orders or local antibiotic policy. So, there is need to develop local and national stewardship program to rationalize and reduce the use of antibiotics for prevention of antimicrobial resistance. There should also be implementation of infection prevention control policies in hospitals to reduce use of antibiotics.

Acknowledgements

The authors acknowledged to Dr Eman Salic, Dr Irfan Ikram, Dr Shivam Thakur, Ms Tania Munawwar, Ms Iffat Latif, Ms Nabiha Shafique, Ms Amna Iftikhar and Ms Maryam Kaleem for taking part in data collection.

Disclosure statement

The authors have no conflict of interest.

Additional information

Funding

The authors have no funding except that access to KOBOCOLLECT software for entering data was provided by World Health Organization.