Impact of Obesity on Echinocandin Effectiveness in Treating Candida Infections: A Retrospective Observational Cohort Study

Introduction

Fungal infections caused by Candida species could lead to invasive infections associated with serious medical complications. These infections are very common in healthcare environments and are considered one of the leading causes of infection-related morbidity and mortality.¹ The cornerstone of the treatment of IC infections is echinocandins including anidulafungin, caspofungin and micafungin.² The dosing regimens of echinocandins approved by the FDA are indicated for adult patients as follows: anidulafungin at a 200 mg loading dose followed by a maintenance dose of 100 mg; caspofungin at a 70 mg loading dose, followed by 50 or 70 mg as maintenance; and micafungin 100 or 150 mg without loading dose. These SDs are fixed across all BMI categories, including morbidly obese patients.³

Obese patients have unique PK parameters that are characterised by alterations in the volume of distribution (VD) and clearance (CL) compared to normal-weight patients.^4,5 These differences in PK parameters have resulted in several PK studies recommending the dose adjustment of echinocandins in obese patients to achieve the required minimum inhibitory concentration (MIC) and to avoid therapeutic failure.^6–8 Furthermore, there have been few studies investigating the clinical outcome variations between obese and non-obese patients receiving the SDs of echinocandins.^9,10 However, these studies have shown conflicting data regarding the clinical outcomes difference in the effectiveness of SDs of echinocandins in treating Candida infections in obese compared to non-obese patients. Therefore, due to the lack of clinical studies in the current literature and conflicting data regarding the effectiveness of SDs of echinocandins in obese patients, our study aims to evaluate the clinical outcomes in obese and non-obese patients who received SDs of one of the echinocandins during treatment for IC.

Materials and Methods

Study Design, Setting, and Patient Population

An observational retrospective cohort study was conducted from the 1^st of January 2017 until the 1^st of February 2023 at KSUMC. The study was approved by the institutional review board at KSUMC on 14/05/2023 with Ref. No. 23/0325/IRB. The study included hospitalised adult obese and non-obese patients who received more than four consecutive doses of SDs of one of the echinocandin agents mentioned previously. Also, only patients with a confirmed diagnosis of Candida infection were included. Diagnosis of Candida infection was confirmed by a positive culture of any type of Candida spp and clear documentation of diagnosis from an infectious disease physician. The study excluded patients with BMI < 18 Kg/m² and those who did not complete the course of the treatment in the hospital. Additionally, Figure 1 shows all the inclusion and exclusion criteria applied during the study period from 2017 – until the 1^st of February 2023.

Figure 1 The patient inclusion process.

Variables and Definitions

Obese patients were defined as patients with BMI ≥ 30 kg/m², based on the definition of the Centers for Disease Control and Prevention (CDC) and World Health Organisation (WHO).^11,12 Demographic and comorbidity variables were collected from the 1^st day of admission. Microbiological data, including the type of Candida spp and type of Candida infections, were collected from the 1^st day of confirmed diagnosis. Subsequently, the types of Candida infections were classified into candidemia and non-candidemia infections, which included intra-abdominal, abdominal, wound, lung and any IC infections other than candidemia. Treatment data, including the name of the medication and its dose, were collected from the date of initiation.

Outcomes and Definitions

The primary outcomes include the total LOS, the duration of Candida infection and the incidence of clinical resolution. We calculated the LOS from the 1^st day of echinocandin treatment until the date of discharge, while the duration of Candida infection was calculated from the date of diagnosis with Candida infection until the last day of echinocandin treatment. In addition, clinical resolution outcomes were determined from a physician’s documentation in the patient’s medical records, indicating no need for continued or additional antifungal therapy due to clinical resolution. The secondary outcome included all-cause mortality in the hospital.

Data Collection and Analysis

We extracted the data from electronic medical records using the Cerner Millennium System (eSIHI) used in KSUMC. Data were collected in Microsoft Excel sheets and appropriately revised. Revised data were then moved to the Statistical Package for Social Sciences (SPSS) version 26 for statistical analysis. In descriptive statistics, normally distributed variables are presented as the mean and standard deviation (SD), while non-normally distributed data are presented as the median and interquartile range (IQR). Categorical variables are described as frequencies and percentages. Associations between categorical variables were tested using Pearson’s Chi-square statistical test. Means of normally distributed scale data between obese and non-obese patients were compared using the independent-samples t-test while non-normally distributed continuous data were tested using the Mann–Whitney U-test. In addition, logistic regression and multivariate analysis were used for primary outcomes, with adjustments made for potential confounders. A P value less than 0.05 was considered to indicate statistical significance.

Results

After screening the medical records by applying the inclusion and exclusion criteria (as shown in Figure 1), we included 132 patients for the final analysis 47 obese and 85 non-obese. The patient characteristics are presented in Table 1. The median age was 61 (IQR 45–68) with no significant difference between the obese and non-obese groups (P= 0.4). There were differences in weight, height and BMI between the groups (P< 0.05). The median BMI was 34.5 (IQR 32–37.40) kg/m² for obese patients and 24 (IQR 21.90–26.50) kg/m² for non-obese patients. In total, 67 (50.8%) of the included patients were male and 65 (49.2%) were female. Male patients were more frequent in the non-obese group (50 (58.8%)) compared with the obese patients group (17 (36.2%)).

Table 1 Basic Characteristics of All Patients with Candida Infections Treated with the SDs of Echinocandins

Additionally, the comorbidity frequency was comparable between the groups, except that immunocompromised and malignant patients were noted more frequently in the non-obese patient group. The type of Candida infections (candidemia vs non-candidemia) did not show a statistically significant difference between the two groups, with C. albicans and C. glabrata being the most common causes of these infections. In total, 63.3% of patients were treated with SDs of micafungin, while 36.4% were treated with caspofungin. The treatment selection was found to be similar for both obese and non-obese patients.

The median length of Candida infections in both obese and non-obese patients was 20 days (IQR 16.25–28.5) and 17 days (IQR 14–30), respectively; with no statistically significant difference between the two groups (P= 0.160). In addition, the median LOS duration was found for obese (29.5 days (IQR 19.25–57.5)) and for non-obese (29.5 days (IQR 18.5–42.25)) patients with no significant differences between the groups (P= 0.896). Clinical improvements among obese and non-obese patients were noted in 32 cases (68.1%) and 56 cases (65.9%) respectively with no significant differences between the two groups (P= 0.797). In addition, the all-cause mortality rates among obese and non-obese patients with Candida infections were 21 (44.7%) and 36 (42.4%), respectively, with a P value of 0.796 (see Table 2).

Table 2 Clinical Outcomes of Obese and Non-Obese Patients with IC Infections Including Candidemia and Non-Candidemia Infections

A subgroup analysis was conducted on patients based on the type of infection: either candidemia or non-candidemia infections. In patients with candidemia (a total of 71 patients; 27 obese and 44 non-obese), no significant difference was observed in the primary and secondary outcomes between the two groups (P> 0.05). In addition, the primary and secondary outcomes presented no significant differences in the groups diagnosed with non-candidemia infections with a P value of > 0.05. (see Table 3). Furthermore, as shown in Table 4, the predictive impact of obesity status on the duration of infection and LOS in patients with invasive Candida infections treated with SDs of caspofungin and micafungin was non-statistically significant. This finding was derived from multivariate linear regression analyses that controlled for age, gender, comorbidities, and the type of echinocandin used. Additionally, logistic regression analyses indicated that obesity status did not a have statistically significant impact on clinical improvement in these patients, even after adjusting for the same potential confounders (see Table 4).

Table 3 Clinical Outcomes of Obese and Non-Obese Patients with Candidemia and Non-Candidemia Infections

Table 4 Regression Analysis of Primary Outcomes with Controlling Potential Confounders

Discussion

In recent years, the rate of obesity has increased significantly, leading clinicians to encounter more obese patients with various serious illnesses. Unfortunately, clinical studies or data on dosing regimens in obese patients for various medications are often limited or absent. This can make it challenging for clinicians to ensure that obese patients receive appropriate therapeutic dose regimens, particularly for serious illnesses, such as IC, including candidemia.¹³

Several PK studies have shown lower exposure to the SDs of echinocandins in obese compared to normal-weight patients, recommending dose adjustment to achieve the required PK/PD targets.^7,8,14–16

Conversely, some studies have found no significant variations in echinocandin exposure in obese compared to normal-weight patients, leading to no dose adjustment recommendations.^17,18 Due to conflicting findings and the limited research evaluating the variations in clinical outcomes between obese and normal-weight patients treated for IC using SDs of echinocandins, our study was conducted to identify any potential differences in the clinical outcomes between these two groups when treated with the SDs of one of the echinocandins.

Our study found no significant differences in clinical outcomes between obese and non-obese patients treated with SDs of caspofungin or micafungin for IC, including candidemia and non-candidemia infections. Primary and secondary outcomes showed no significant differences, aligning with Results from other studies.

In our findings, all-cause mortality rates were not significantly different between obese and non-obese patients with candidemia. Similar results were noted by Hutton et al (2022) in their study on SDs of anidulafungin, where no significant difference was found in the 30-day all-cause mortality among BMI categories (P= 0.976). Additionally, Barber et al (2020) conducted a study that found no differences in hospital mortality between obese and non-obese groups with candidemia receiving the SDs of micafungin (P= 0.36).

In addition, Hutton et al (2022) found no significant differences in the clinical response across BMI categories in patients who received SDs of anidulafungin to treat candidemia. These results are in accordance with our findings, which showed no difference in the clinical improvement outcomes between obese and non-obese patients with candidemia infections (P= 085). Furthermore, a post hoc analysis of nine clinical trials on caspofungin’s SDs regimen showed similar clinical success rates (71% to 77%) across BMI categories.¹³ Our study also observed similar clinical improvement percentages among our included both obese and non-obese patients (60%, 66.6% and 73% in patients with candidemia, IC, and non-candidemia infections, respectively). Notably, these results, showing the effectiveness of SDs of echinocandins across various BMI categories, support other findings that showed successful treatment in a critically ill obese patient (BMI >40 kg/m²) treated with SDs of micafungin for a urinary tract infection, despite the lower serum drug concentration noted in this patient.¹⁹

Moreover, our study found a total median LOS of 29 days for IC patients, consistent with a similar study on obese patients with IC treated with a high dose of micafungin (around 300 mg daily, median BMI 37 kg/m²) in which the median LOS was 27 days.²⁰ Despite the high dose used in this study, the duration of total LOS is comparable with our findings for IC patients treated with SDs of either micafungin or caspofungin.

However, a study conducted in 2020 to evaluate the impact of obesity on candidemia patients treated with micafungin, fluconazole and posaconazole, found that the obese group had a longer infection-related LOS by 7 days compared to the non-obese group.¹⁰ In contrast, our study found no significant difference in the total hospital LOS between the two groups (P= 0.889). This difference in LOS between our study and that of Barber et al (2020) could be due to differences in the study population; our study included only those treated with the SDs of echinocandins, while theirs included patients treated with micafungin, representing 73% of the total patients, with the others receiving either fluconazole or posaconazole.

Additionally, Barber et al (2020), found a significant difference in the median duration of candidemia resolution between obese and non-obese groups (P= 0.02), whereas our study showed no significant difference (P= 0.226). However, this variation in results may be attributed to differences in how the resolution duration was calculated; Barber et al (2020) counted from the first positive to the first negative culture, while we calculated from the first confirmed diagnosis day until the discontinuation of medication due to clinical resolution.

Moreover, our study identified C. albicans and C. glabrata as the most prevalent Candida types causing IC, including both candidemia and non-candidemia infections. This aligns with findings from other epidemiological studies, showing that these two Candida pathogens are commonly associated with most IC infections.^21–23

Nevertheless, retrospective studies are susceptible to selection bias and confounder effects. In this study, efforts to mitigate bias included strict adherence to inclusion and exclusion criteria for both obese and non-obese patients, consistent application of clear outcomes and variable definitions for both groups and ensuring comparable data between the two groups. However, no study is without limitations, and our study has some that should be taken into account. Firstly, the weight distribution among the obese patients included in our study may not represent the broader obese population due to the limited range of obese and morbidly obese patients included in our study. Thus, further studies to include large numbers of patients with high and extremely high weights are required. Secondly, the retrospective single-centre design may have limited the sample size and, therefore, the generalisability of the results. To overcome such limitations, future studies should be conducted in multicentre settings in order to generate a more representative sample and more generalisable results.

Conclusion

This research study examined the relationship between obesity and the effectiveness of SDs of echinocandins in treating IC. No significant differences were found in clinical outcomes between the obese and non-obese patients, indicating that factors such as BMI and its related PK variations may not significantly affect the therapeutic efficacy of echinocandins in particular caspofungin and micafungin. However, further research is required to investigate the clinical outcomes of SDs of echinocandins in obese patients to ensure safe pharmacotherapy.