Real-World Effectiveness of Meropenem: Results of an Observational Study in 84 Colombian Patients

Background. To determine the effectiveness and safety of meropenem in routine clinical practice, in terms of clinical and microbiological response. Methods. A real-world, observational, descriptive, longitudinal study with daily monitoring of clinical history records was conducted on all patients who were medically prescribed meropenem during a period between October 2015 and March 2016 at a university hospital in Bucaramanga (Colombia). Results. The study evaluated 84 patients with an average age of 63.2 years, mostly older adults with multiple comorbidities, of whom 54.8% were men. A positive clinical or microbiological response was obtained in 98.8% of the patients. At the end of the treatments, significant improvements in dysthermia (0% vs 29% at the beginning, p = 0.000), tachycardia (13% vs 47%, p = 0.049) and leukocytosis (39% vs 15% at the beginning, p = 0.008) were evidenced. The improvement in the indicator that combines all the SIRS criteria was also significant (p = 0.000). The treatment was well tolerated, although we identified some non-serious and expected adverse reactions. Conclusions. Generic meropenem proved to be effective and well tolerated for different types of infection in routine clinical practice. The results are consistent with the findings of the clinical studies with the innovator drug.


Introduction
Antimicrobials have been of great therapeutic use for more than half a century as indispensable tools for treating a wide variety of infections [1,2], making it possible to save the lives of millions of people. Meropenem is a widely used antimicrobial in the treatment of many complicated infections. Its wide spectrum, its powerful effect on microorganisms with different resistance mechanisms and its good penetration of tissues make it a highly useful part of the therapeutic arsenal [3][4][5], making it necessary to monitor its use and results at a local level in order to conserve its effectiveness and safety in terms of the development of bacterial resistance and other related problems. In this sense, real-world studies are becoming increasingly important as a trustworthy source of evidence for the behaviour of treatments in routine clinical practice and as supports for clinical decision making. Furthermore, these studies include patients that are normally excluded from clinical studies and provide information about local use of medications, which makes it possible to identify factors that can have an impact not only regarding safety but also in terms of therapeutic results [6,7].
The effectiveness of different generic antimicrobials has previously been studied in different situations [8][9][10]. At a local level, there is evidence for its equivalence to innovative medications in terms of mortality, length of hospital stay and safety [11]. Studies from around the world have described positive results for meropenem in terms of its effectiveness and tolerability both in complicated infections and in other important conditions [12][13][14][15]. However, local evidence is limited and there is a need for real-world studies that can support decision making. This study sought to address this need by determining the effectiveness of this carbapenem in terms of the clinical and microbiological response in patients treated in a six-month period at a university hospital in Colombia.
In most patients, the dose used was 1 gram every 8 hours (n = 72, 85.7%). Other prescription regimes are described in Table 2. The specialty that most prescribed meropenem was internal medicine (n = 45, 53.6%). The median treatment duration was 9 days (IQR 7-12). At the end of the meropenem treatment, improvement was observed in all SIRS criteria individually. Qualitatively, the response was significant for dysthermia, tachycardia and leukocytosis (Table 3), and quantitatively in the four variables: body temperature (p = 0.000), breathing rate (p = 0.018), heart rate (p = 0.000) and leukocyte count (p = 0.000). The proportion of patients with SIRS at the end of treatment was significantly lower than at the beginning (p = 0.05) (Figure 2). The aggregate response of the SIRS criteria through the combined indicator was also significant (p = 0.000 for difference of means, p = 0.000 for difference of variance).

Figure 2. Proportion of patients with SIRS over time treatment
The therapeutic objective was achieved in 83 of the 84 treatments analysed (98.8%). Most of the patients presented clinical response (n = 82, 97.6%), while complete response was evident in 23.8% of the cases (Table 4). During the monitoring, we identified some patients who presented non-serious and expected adverse reactions, such as skin manifestations of hypersensitivity (n = 6, 7.1%), constipation (n = 4, 4.8%) and a patient with oral candidiasis (n = 1, 1.2%). We found no clinical or paraclinical evidence of potential interactions of meropenem with other drugs.

Discussion
Different clinical studies with meropenem show efficacy between 64% and 99%, depending on the type of infection. The highest success rate has been achieved in urinary tract infections (90.0% -99.0%), followed by skin and soft tissue infections (73.1% -98.0%) and lower respiratory tract infections (64.0% -89.0%) [3,4]. Meanwhile, real-world studies have shown that the effectiveness of generic meropenem has ranged from 72.5% to 80.0% [12][13][14][15]. In the present study, the effectiveness was determined to be 98.8%, which is within the range reported in clinical studies and is higher than in other real-world studies. This may be because patients with inadequate therapy were not included in this study, i.e. those cases where non-meropenem-susceptible microorganisms were identified.
The effectiveness of meropenem in this study was determined by qualitative and quantitative methods, both for each SIRS criterion individually and for all of them together. Based on these indicators, the disappearance of SIRS over time was evidenced for a significant proportion of patients. These results were obtained despite the complexity of the patients and other factors that threaten therapeutic success, including older patients with multiple comorbidities such as hypertension, diabetes, COPD, kidney disease and HIV. These baseline conditions are associated with various complications, including increased hospital stay and mortality in respiratory infections [16][17][18], multi-resistance in urinary pathogens [19][20][21], and worse prognosis in bacteremia [22].
The case of therapeutic failure corresponds to an 82-year-old patient with congestive heart failure, stage III chronic renal disease, hypothyroidism and COPD, treated in the intensive care unit for cardiogenic shock, atrial fibrillation and respiratory failure requiring invasive mechanical ventilatory support. In addition, there was suspected myelodysplastic syndrome due to bicytopenia, which progressed to pancytopenia. The patient presented bronchoaspiration pneumonia, for which they received meropenem for 10 days. They went on to suffer from dyspnea, tachycardia, subsequent multiorgan failure, and finally died. The expected mortality according to APACHE II was 40%, rising to 80% due to comorbidities, despite optimal treatment. At the end of the treatment with meropenem, Stenotrophomona maltophilia was isolated in tracheal aspirate. This bacterium is an opportunistic microorganism that, as it possesses intrinsic resistance to several antimicrobials, could be associated with the outcome. It is inducible for L1 beta-lactamases with class B zinc enzymes, highly active against carbapenems, and by protocol it is only tested by antibiogram against trimetroprim/sulfamethoxazole, the treatment of choice [23].
Cultures were taken in only two thirds of the patients, which could be due to the difficulty of obtaining samples in cases of pneumonia. A control culture was performed in less than half of the cases with an identified culture. Regarding the isolated microorganisms, we mainly identified pathogens that have developed significant antimicrobial resistance: Pseudomonas aeruginosa, Escherichia coli and Klebsiella pneumoniae.
The most frequently used dosage was 1 gram every 8 hours, in accordance with the dosage recommendations established for this carbapenem. Dose adjustments were identified in cases of renal failure (500 mg every 12 hours and 1 gram every 24 hours). In some cases of bacteria with increased minimum inhibitory concentration, doses of 2 grams every 8 hours were used. In three cases, the treatment lasted more than 40 days, two of which corresponded to osteoarticular infections and another to a case of perianal abscesses that were difficult to treat.
As an important contribution regarding the rational use of meropenem, in 18 patients who received initial empirical treatment with meropenem, the treating physicians decided to de-escalate based on the result of the antibiogram. The use of this carbapenem as initial therapy in the emergency department by general practitioners is also noteworthy; however, we confirmed that these were professionals trained by the urology department who started the treatment following the criteria established by this institution.
Around the world, it has become evident that there is a need to strengthen medical education by avoiding the prescription of antimicrobials in cases that do not warrant it. In this study, cases without indication for antimicrobial use were identified, including asymptomatic bacteriuria and colonization [24]. Another issue to highlight is that no adjustments were seen in the administration Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 19 November 2020 doi:10.20944/preprints202011.0512.v1 of the drug in haemodialysis patients, bearing in mind that up to 50% of meropenem is removed during this process, and so doses are required after the sessions [25,26]. These considerations for usual practice can be modified in order to prevent the development of bacterial resistance, in this case carbapenemase production. In terms of safety and tolerability, the adverse reactions identified were non-serious, expected, and their frequency is within the expected range [4,5]. Some analyses indicate that cross-reactivity between penicillin and meropenem is low (0.06% -1.9%); however vigilance and caution are recommended [27]. Two of the seven patients with previous hypersensitivity to penicillins presented cutaneous manifestations. The first one was given a dose of piperacillin/tazobactam before increasing the dose of meropenem, and the other patient received six concomitant drugs, so the strength of association with meropenem is reduced, but they are still plausible cases. Another possible adverse event of special interest in terms of safety is the lowering of the seizure threshold when using carbapenems [5]. Despite observing patients with risk factors, this potential reaction was not identified in any case.
In the context of decision-making and efficiency in health spending, the WHO has urged developing countries to promote the use of quality generic medicines. Some animal models have suggested that the behaviour of the generic drug is not equivalent to that of the innovator [28]; however, they have been questioned as they have not been validated to assess antimicrobial efficacy [8]. This study shows generic meropenem to be effective and well tolerated for different types of infection in routine clinical practice.
The strengths of this study include the precision in obtaining the data, the frequent monitoring, having considered 100% of the patients eligible (thereby guaranteeing representativeness in those risk groups that are usually excluded from clinical studies) and not considering those cases in which there is no evidence of infection, even though they could have been considered successful examples of bacterial eradication in cases such as asymptomatic bacteriuria. The limitations include the small number of patients, implementation at a single hospital centre, and non-inclusion of outcomes such as hospital stay and mortality.
In future research, it is recommended to evaluate compliance with institutional antibiotic use guidelines and the ways the medication is used by the nursing group, including reconstitution and dilution technique, infusion time and compliance with the timing of antimicrobial administration. Likewise, it is important to make an in-depth evaluation of pharmacokinetic aspects of interest such as renal failure, and their respective dose adjustments, correlating these variables with the outcome.
With these findings, it can be concluded that treatments with meropenem (Vitalis) in the usual clinical practice proved to be effective and well tolerated, with results that are within those expected according to previous data in clinical and real-world studies. As regards opportunities for improvement in the use of meropenem, several factors were identified that can be managed to mitigate the risk of failure and generation of bacterial resistance.

Materials and Methods
We carried out a real-life, descriptive, longitudinal and prospective observational study at a tertiary university hospital in the city of Bucaramanga (Colombia), through the daily monitoring of all the consecutive patients who were medically prescribed meropenem, in a six-month period between October 2015 and March 2016.
Medical records were reviewed, and assessments performed without any interventions or modification of current therapies. This study followed the Good Clinical Practices for Investigation in humans established in the Declaration of Helsinki and the 1993 Resolution 8430 of the Colombian Ministry of Health.
The confidentiality of the information was ensured. The study was approved by the Ethics Committee of Research of the the university hospital where the study was carried out.
We included the patients of all ages, degrees of severity, comorbidities, types of infection, who were given meropenem (Vitalis in Colombia) regardless of the services and speciality of the institution that referred them. In order to characterize the patients, we collected demographic, clinical, laboratory, pharmacological and microbiological variables based on their clinical history records in a case report form (CRF). The information was collected and collated by a trained professional in a previously tested and adjusted database (Microsoft Excel). The data obtained were validated by the researchers.
The following were considered as exclusion criteria: a) non-compliance with the criteria defined in the reference guide b) if the microorganism causing the infection was identified as being resistant to or not susceptible to meropenem, c) de-escalation based on an antibiogram for meropenem, d) the patient was treated for less than 72 hours, and e) if the treatment began in another institution.
Effectiveness was defined as those cases which had a clinical response, microbiological response, or both. Cases which had both clinical and microbiological improvement were classified as having a complete response. Clinical response was defined as a) explicit evidence of improvement in the clinical history in the opinion of the treating physicians, or b) evidence of disappearance of systemic inflammatory response syndrome (SIRS) following treatment [37,38]. In applicable cases, the treatment includes surgery, in order to control the source of the problem (e.g. infected diabetic foot). Microbiological response was defined as the neutralisation of the control culture (eradication of the etiological agent) in the cases in which this analysis was made. Therapeutic failure corresponded to those cases in which there was no clinical or microbiological response [39].
This study was approved by the Research Ethics Committee of the University Hospital as risk-free research (resolution 8430, Ministry of Health, Colombia), respecting the principles of the Declaration of Helsinki. The committee did not consider the need for informed consent, given the nature of the study. The observation of the treatments began after meropenem was prescribed according to the criteria of the treating physicians.
The data were analysed using the free software R (The R Foundation). Results were expressed by univariate analysis in percentage frequencies, means with standard deviations and medians with interquartile ranges. Bivariate analyses were performed to study the change in each SIRS criterion between the beginning and end of treatment, and likewise the change in the SIRS as a whole. In qualitative analyses (yes / no) we used chi-square and Fisher's exact tests, and for quantitative analyses we used the t-test for difference of means or the Wilcoxon test for non-normal distributions. Additionally, a unified indicator was created by means of standardization to evaluate the response of all the SIRS variables before and after the treatment. In all cases, the level of statistical significance was set as p ≤ 0.05.