Burden of serious fungal infections in the Netherlands

Summary Background Fungal diseases have an ever‐increasing global disease burden, although regional estimates for specific fungal diseases are often unavailable or dispersed. Objectives Here, we report the current annual burden of life‐threatening and debilitating fungal diseases in the Netherlands. Methods The most recent available epidemiological data, reported incidence and prevalence of fungal diseases were used for calculations. Results Overall, we estimate that the annual burden of serious invasive fungal infections in the Netherlands totals 3 185 patients, including extrapulmonary or disseminated cryptococcosis (n = 9), pneumocystis pneumonia (n = 740), invasive aspergillosis (n = 1 283), chronic pulmonary aspergillosis (n = 257), invasive Candida infections (n = 684), mucormycosis (n = 15) and Fusarium keratitis (n = 8). Adding the prevalence of recurrent vulvo‐vaginal candidiasis (n = 220 043), allergic bronchopulmonary aspergillosis (n = 13 568) and severe asthma with fungal sensitisation (n = 17 695), the total debilitating burden of fungal disease in the Netherlands is 254 491 patients yearly, approximately 1.5% of the country's population. Conclusion We estimated the annual burden of serious fungal infections in the Netherlands at 1.5% of the population based on previously reported modelling of fungal rates for specific populations at risk. With emerging new risk groups and increasing reports on antifungal resistance, surveillance programmes are warranted to obtain more accurate estimates of fungal disease epidemiology and associated morbidity and mortality.


| INTRODUC TI ON
Fungal diseases have a significant impact on human health and affect over 1 billion people worldwide. 1 Disease burden ranges in severity from mild skin, nail and hair infections affecting a high proportion of the global population, to severe invasive fungal infections affecting a smaller immunocompromised population. Current national estimates of more severe invasive fungal infections are lacking but experts believe that the global rate of invasive fungal infections is increasing due to an increasing population receiving immunosuppressive therapy, a growing elderly population and increased survival from previously lethal diseases. Epidemiological knowledge on fungal infections in the Netherlands is limited due to tenuous hospital and national surveillance data. In this study, we estimate the burden of serious (life-threatening) fungal infections in the Netherlands, in concordance with the globally initiated LIFE programme (www.LIFEworld wide.org).

| MATERIAL S AND ME THODS
The annual burden of fungal infections in the Netherlands was assessed using fungal infections frequencies for chronic pulmonary aspergillosis (CPA), allergic bronchopulmonary aspergillosis (ABPA), severe asthma with fungal sensitisation (SAFS) and invasive aspergillosis, cryptococcosis, Pneumocystis pneumonia, mucormycosis, Fusarium keratitis, candidemia, Candida peritonitis and vulvo-vaginal candidiasis. A literature search was conducted in order to identify specific fungal infection incidences and epidemiological reports from the Netherlands. When epidemiological data was unavailable, the incidence or prevalence of fungal infections was estimated based on the specific population at risk and the reported incidences for these risk groups. For all estimates, the most recent epidemiological data available was obtained for this study.
The authors confirm that the ethical policies of the journal, as noted on the journal's author guidelines page, have been adhered to. No ethical approval was required as this is a review article with no original research data.
Population data for 2017 were obtained from Statistics Netherlands (CBS). 2 The total population in the Netherlands in 2017 was 17 081 507. The female population between 16 and 50 years totalled 3 667 386. The current total patients diagnosed with HIV in the Netherlands (n = 20 800), as well as the proportion of confirmed cases on anti-retroviral treatment (93%, n = 19 289), and the annual new AIDS cases (n = 724) were retrieved from the HIV monitoring report 2018. 3 The annual incidence of patients with tuberculosis (n = 484; with and without HIV coinfection) was obtained from a surveillance report on tuberculosis in the Netherlands 2016. 4 The prevalence of COPD in 2018 (n = 613 800, all stages of disease), as well as the number of hospital admissions in 2017 (n = 33 735) and the total number of patients with asthma in 2018 (n = 636 200), was obtained from the 'public health and healthcare' website. 5,6 The number of children with asthma was estimated to be approximately 100 000. The number of patients with cystic fibrosis in 2017 (n = 905) was retrieved from the ECFS Patient Registry Annual Data Report. 7 The yearly number of solid organ transplants (n = 1292, of which 998 were renal) was obtained from the 'Dutch transplant foundation' website, 8  it is estimated that 1/3 of allogenic HSCTs were from related donors and twice as many autologous HSCTs were performed. Therefore, we used a yearly number of 611 allogeneic and 1,222 autologous HSCTs in the model. The yearly number of patients with acute myeloid leukaemia (AML) (n = 732) was retrieved from the national cancer registry, 9 where the average of 3 years (2015-2017) was used to estimate the average number of patients. Total number of intensive care unit (ICU) admissions for 2017 (n = 80 687) was obtained from the 'National Intensive Care Evaluation' (NICE) foundation. 10 The rate of invasive aspergillosis was estimated to be 10% for AML patients, 0.5% in renal transplant recipients, 4% for lung and liver transplants and 6% for heart transplants, based on an epidemiological study from France. 11 The percentage of invasive aspergillosis in non-AML patients with haematological disease was estimated to be similar to that for AML patients, including autologous transplantation. 11,12 The rate of invasive aspergillosis in patients with influenza admitted to the ICU was estimated based on the number of influenza associated pulmonary aspergillosis (IAPA) cases reported for the 2015-2016 season in Dutch University Medical Centres. A total of 144 patients with influenza were admitted to the ICU and 16% of these patients matched the definition of IAPA. 13 The ICU beds in the beforementioned Dutch University Medical Centres cover 22% of the total ICU beds in the Netherlands. 14 The incidence of IAPA in ICU beds in University Medical Centres was extrapolated to the total number of ICU beds in the Netherlands. We estimated that approximately 50% (45/83 or 54%) 15 of patients with IAPA were immunocompetent and therefore to prevent counting duplicates, only these non-immunocompromised patients were considered additional cases of invasive aspergillosis. To calculate the rate of azole resistance in invasive aspergillosis, we used the mean resistance percentage (11.3%) based on a national surveillance programme of five University Medical Centres (14.7%) and five teaching hospitals (7.8%). 16 The incidence of mucormycosis was based on an estimated population incidence of 0.09 per 100 000 population per year (averaged over 10 years). 17,18 The incidence of Fusarium keratitis was calculated using the mean incidence of 0.045 per 100 000 population per year between 2005 and 2016 reported in a recent retrospective study on Fusarium keratitis in the Netherlands. 19 The annual and 5-year prevalence of CPA in patients with tuberculosis was based on a method described previously. 20 Tuberculosis as the underlying cause of CPA was predicted to be 25%. The number of ABPA cases in asthma was estimated to be 2.5%, 21 and the burden of ABPA cases in cystic fibrosis (CF) was estimated to be 18%. 22,23 The burden of SAFS was estimated to be 33% of the 10% most severe asthmatic adult patients. 24 The number of invasive aspergillosis cases in COPD patients was estimated based on a previously reported frequency of invasive aspergillosis in admitted COPD-patients of 1.3%. 25 The annual incidence of candidemia for the Netherlands is calculated to be 2.61 per 100,000 population. In Europe 26 and globally, 27 5.52 and 6.87 episodes per 1000 ICU admissions are diagnosed with Candida bloodstream infections, respectively. Candida peritonitis (intra-abdominal candidiasis) was set to the European average of 1.84 episodes per 1000 ICU patients. 26 We assumed that 6% of women aged 16-50 had recurrent vulvo-vaginal candidiasis. 28

| RE SULTS
We estimated that the annual burden of serious invasive fungal infections in the Netherlands was 3185 while the total number of debilitating fungal infections was 254 491 (Table 1)  The rate of ABPA in adults with severe asthma and CF was estimated to be 13 568 while the burden of SAFS was 17 695. We reduced the total number of patients with ABPA and SAFS for calculation of the fungal burden by 25% (7816 patients) as some patients with ABPA may also have SAFS. 35 Altogether, the annual debilitating burden of fungal diseases in the Netherlands was approximately 1.5% of the country's population. The total number of patients with ABPA and SAFS was reduced with 25% for the calculation of the total fungal burden to prevent duplicates. c Total serious fungal infection burden and Total fungal burden are the sum of the above.  13,15 In addition, other diseases may temporarily increase the incidence of invasive aspergillosis.

| D ISCUSS I ON
There are indications that critically ill patients with SARS-CoV-2 may be at risk to develop invasive aspergillosis. Li and Xia described CT findings in patients with SARS-CoV-2 suggestive of invasive aspergillosis including halo or reversed-halo sign. 36 However, as data on invasive aspergillosis in COVID-19 patients is currently sparse, it remains uncertain whether the risk of invasive aspergillosis in critically ill patients with SARS-CoV-2 exceeds the general risk of invasive aspergillosis in critically ill patients. In total, we estimated the rate of invasive aspergillosis to be 7.7/100 000, which is somewhat higher to that reported by surrounding countries. Belgium estimated the rate to be 6.08/100 000, while the rate in Germany and Denmark was 5.1/100 000. 30,37,38 Largely, we used the same methodology for calculating the invasive aspergillosis infection rate. However, we have added the number of patients with IAPA while Belgium, Germany and Denmark did not. Regardless, the number of patients with IAPA only accounted for a small proportion of the higher rate (0.33/100 000). The higher estimated rate can be explained by our relatively high number of transplantation and COPD patients. Some risk groups for invasive aspergillosis like lung cancer, which has an incidence of up to 2.6%, were not taken into account. 39 In addition, we used a frequency of invasive aspergillosis in admitted COPDpatients of 1.3% which was based on study from Spain. 25 Another study from China reported a frequency of invasive aspergillosis of 3.9% in admitted COPD-patients. 40 The actual number of patients with invasive aspergillosis may be surpassed due to an ever increasing number of solid and hematopoietic stem cell transplantations that are treated with novel treatments strategies with a shorter time to neutrophil recovery.
Here, to the best of our knowledge, we have made the first estimate on the rate of azole resistant invasive aspergillosis. The azole resistance rate in invasive aspergillosis was based on a national surveillance programme incorporating data from five University Medical Centres and five teaching hospitals in 2018. 16 We used the mean resistance frequency from these hospitals to extrapolate the azole resistance frequency in the Netherlands. We calculated that 145 patients (11.3%) with invasive aspergillosis may be infected with an azole resistant isolate. This is a significant finding as patients with voriconazole resistant invasive aspergillosis had a 20%-25% hence, our calculations may underestimate the actual burden of CPA. 43,44 The estimated annual incidence of Fusarium keratitis is 8. However, this estimate is based on the mean incidence between 2005 and 2016. Interestingly, the incidence of Fusarium keratitis has been increasing over the years and 15-25 annual cases were found in 2014 to 2016. 19 Similar trends are seen in Germany. 45 Thus, the estimated incidence of 8 may underestimate the current incidence of Fusarium keratitis.

ACK N OWLED G M ENTS
This research received no external funding

CO N FLI C T S O F I NTE R E S T
The authors declare no conflict of interest.