Peritonitis Due to Thermoascus taitungiacus (Anamorph Paecilomyces taitungiacus)

Abstract

The first case of human disease due to the thermophilic ascomycete Thermoascus taitungiacus (the teleomorph of Paecilomyces taitungiacus) is presented. T. taitungiacus was recovered from four dialysate fluid specimens of a 57-year-old patient undergoing chronic peritoneal dialysis. Identification was based upon cylindrical conidia, reddish orange nonostiolate ascomata, lack of growth at 20°C, thermotolerance, and ascospores that appeared pale yellow, elliptical, thick walled, and predominately echinulate by light microscopy but irregularly verrucose by scanning electron microscopy.

Paecilomyces species are found worldwide in soil, water, and decaying vegetation (13, 29) and are microscopically similar to Penicillium and some Aspergillus species. While they are uncommon human pathogens, they have been associated with serious infections in both immunosuppressed and immunocompetent patients, especially those with defects in the anatomic barriers or with a foreign body. Five species in the genus Paecilomyces have been reported to cause infection, including Paecilomyces variotii, Paecilomyces lilacinus, Paecilomyces marquandii, Paecilomyces viridis, and Paecilomyces javanicus (13). The most commonly reported infections are keratitis, endophthalmitis, and cutaneous infections. Fungemia, prosthetic valve endocarditis, lung infections, sinusitis, and peritonitis are less frequently reported (11, 20, 23, 24, 26, 29). Thermoascus crustaceus has been reported from monocyte cultures of patients with AIDS (14). We report what we believe to be the first case of human mycosis due to the ascomycetous fungus Thermoascus taitungiacus causing peritonitis in a patient undergoing chronic peritoneal dialysis (CPD).

Case report.

A 57-year-old man presented on 25 October 1998 with a 5-day history of abdominal pain. Since 1994 the patient had been treated with CPD for end-stage renal failure due to chronic glomerulonephritis and malignant hypertension. He had had two episodes of bacterial peritonitis in 1996. Physical examination on admission revealed diffuse abdominal tenderness compatible with peritonitis. Empirical intraperitoneal treatment with cefazolin, 0.5 g/2 liters, and aztreonam, 0.6 g/2 liters, was started following a cell count of the turbid dialysate effluent that revealed 250 white blood cells/mm³ (35% neutrophils, 10% lymphocytes, and 55% monocytes). Three additional dialysate fluids were collected on October 27 and 29 and November 1. Each was plated onto blood, MacConkey, chocolate, and Sabouraud dextrose agar (SDA) plates (Hy Laboratories, Rehovot, Israel), as well as Löwenstein-Jensen medium agar tubes (Heipha Diagnostika, Heidelberg, Germany), and incubated at 35°C. All bacterial cultures were negative, but by early November a tan-colored mould grew in pure culture on the Sabouraud agar plates from the four separate dialysate fluid specimens. Intravenous amphotericin B (40 mg/day) was started, and the indwelling Tenckhoff catheter was removed. Hemodialysis was thereafter maintained via a temporary central vein catheter.

Amphotericin B was discontinued after 2 weeks of therapy due to intolerance (severe and repeated vomiting), liver toxicity (elevated serum alkaline phosphatase levels), and the development of a deep vein thrombosis of the upper limb. Oral ketoconazole (200 mg/day) was then given for 2 weeks but was stopped due to the appearance of fever, the accumulation of ascitic fluids, and worsening abdominal pain. The patient became severely catabolic with an accompanying weight loss of 5 kg and hypoalbuminemia of 24 g/liter. Amphotericin B therapy was restarted on December 6 and was given thrice weekly at the end of each hemodialysis session together with intradialytic parenteral nutrition. Fever abated and abdominal pain improved; however, no improvement was seen in either the ascitic fluid volume or the ascitic white cell count. On 20 January 1999, oral itraconazole (400 mg/day) was substituted for amphotericin B, resulting in a gradual decrease in the accumulation of ascitic fluid and complete disappearance of the abdominal pain. The patient started to gain weight, and his serum albumin level increased to 37 g/liter. Itraconazole was discontinued after 5 months of therapy. Repeated dialysate fluid cultures since November 1 were all sterile.

In July 1999, one month after the discontinuation of itraconazole therapy, an enlarging right kidney mass, diagnosed histologically as renal cell carcinoma, was removed at nephrectomy. Two years later the patient is doing well on hemodialysis and remains asymptomatic.

Mycologic studies.

The mould isolated from the peritoneal fluids was initially identified as a Paecilomyces species at the Mycology Unit, Chaim Sheba Medical Center, Tel-Hashomer, Israel. It was then referred to the Fungus Testing Laboratory, Department of Pathology, University of Texas Health Science Center at San Antonio, San Antonio, Texas, where the final identification of T. taitungiacus was made and susceptibility tests were performed. The isolate was accessioned as UTHSC R-3084, subcultured onto in-house-made (21) potato flakes agar (PFA) and SDA (Remel, Lenexa, Kans.), and incubated at 25, 35, and 42°C. A slide culture was also prepared on PFA and incubated at 25°C. Growth of the isolate on PFA (Fig. 1A) and SDA revealed colonies that were initially flat and buff colored but that quickly (within 6 days) became yellowish orange with a brownish yellow reverse. Sparse conidial structures of the Paecilomyces anamorph were evident between 3 and 4 days at 25°C on PFA. Conidia were initially cylindrical to rectangular, approximately 7 to 7.8 by 2.9 to 3.9 μm, and then became elliptical to subglobose (Fig. 1C). With the formation of a Paecilomyces anamorph, additional subcultures onto Czapek Dox and malt extract agar (MEA) (Remel) at 35°C were made in an attempt to induce ascospore formation. Colonies on MEA became reddish orange and granular to crust-like with a brownish orange reverse (Fig. 1B). Globose, nonostiolate ascomata occurred within 7 days on Czapek Dox agar (Fig. 1D) and contained asci that were approximately 15.6 by 11.7 μm (Fig. 1E). Ascospores were elliptical, pale yellow, thick walled, approximately 4 to 5 by 6.8 μm, and predominately echinulate by light microscopy (Fig. 1F). Ascospores were subsequently examined by scanning electron microscopy (SEM). Briefly, stubs were mounted with Spot-O-Glue Avery labels (Diamond Bar, Calif.) and touched to the fungal culture. Preparations were coated with gold-palladium using a Denton Bench Top Turbo III vacuum evaporator (Morristown, N.J.) and were examined with a LEO 435 VP digital scanning electron microscope (Thornwood, N.Y.). When viewed by scanning electron microscope, ascospores appeared irregularly verrucose (Fig. 1G).

FIG. 1.

Macroscopic and microscopic features of T. taitungiacus (anamorph P. taitungiacus). (A) Colonial morphology of T. taitungiacus after 7 days of incubation at 35°C on PFA. Magnification, ×920. (B) Colonial morphology of T. taitungiacus after 7 days of incubation at 35°C on Czapek Dox and malt extract agar. Magnification, ×920. (C) P. taitungiacus anamorph from slide culture on PFA after 7 days of incubation at 25°C on PFA. Magnification, ×920. (D) Nonostiolate ascoma of T. taitungiacus. Magnification, ×230. (E) Asci containing ascospores of T. taitungiacus. Magnification, ×920. (F) Predominately echinulate ascospores of T. taitungiacus viewed by light microscopy. Magnification, ×920. (G) Irregularly verrucose ascospores of T. taitungiacus viewed by scanning electron microscopy. Magnification, ×3,000.

Temperature studies were also performed by inoculating five plates of MEA (Remel) in triplicate with a 1-mm portion of the case isolate. Plates were incubated at 18 to 20, 25, 30, 35, and 42°C for 6 days, and zone sizes were measured with calipers. Temperature studies performed on day 6 indicated mean colony diameters at 20, 25, 30, 35, and 42°C of 0, 16, 85, 85, and 60 mm, respectively. The most abundant ascocarp formation occurred at 30 and 35°C, while maximum conidial production occurred at 42°C.

Susceptibility studies.

The case isolate and the P. variotii control strain UTHSC 90-459 were evaluated for in vitro antifungal susceptibility by the National Committee for Clinical Laboratory Standards broth macrodilution method M27-A (19) modified for mould testing (9, 10). Results indicated 24- and 48-h MICs of amphotericin B of 0.25 μg/ml. Itraconazole 24- and 48-h MICs were <0.015 and 0.06 μg/ml, respectively. Based upon achievable drug concentrations using standard dosing regimens, the isolate appeared susceptible, in vitro, to both antifungal agents tested.

Paecilomyces species lacking known teleomorphs that have been cited as human etiologic agents include P. variotii, P. lilacinus, P. marquandii, P. viridis, and P. javanicus (13). Homothallic Paecilomyces species are reported by their teleomorph name, Thermoascus (2, 5, 14, 25, 27, 28). As the name implies, these are thermophilic or at least thermotolerant fungi. Their identification in clinical laboratories is based upon growth characteristics, temperature studies, and the microscopic morphology of both the anamorph and teleomorph forms, i.e., conidiogenous cells and conidia in the genus Paecilomyces, and ascomata, asci, and ascospores in the genus Thermoascus. The case isolate was initially identified in San Antonio as T. crustaceus based upon its yellow-orange to reddish orange colonies, the Paecilomyces anamorph displaying cylindrical to rectangular-shaped conidia, growth at 25, 30, 35, and 42°C, and nonostiolate, reddish yellow ascomata producing thick-walled, slightly roughened ascospores as viewed by light microscopy. Ascospores measured 6.5 to 8 by 4.5 to 5 μm. However, in preparation for submission of this paper, SEM studies of the ascospores did not match those reported for T. crustaceus in that they were irregularly verrucose (Fig. 1G) rather than finely echinulate. Recently, Chen and Chen (5) reported a new thermophilic species of Thermoascus that failed to grow at 20°C and produced ascospores that were echinulate by light microscopy but irregularly verrucose by SEM. The irregularly verrucose ascospores in the case isolate matched those described by Chen and Chen (5) for T. taitungiacus. An identification of T. crustaceus was eliminated due to the ascospore morphology and lack of growth at 20°C (5). Other Thermoascus species excluded from consideration were Thermoascus aegyptiacus with slightly verrucose ascospores and no growth at 20°C (27), Thermoascus aurantiacus that frequently lacks an anamorphic state, has minutely verrucose ascospores, and fails to grow at 30°C (5, 28), and Thermoascus thermophilus with nearly smooth ascospores and a Polypaecilium anamorph (2). Features of thermophilic Thermoascus species are displayed in Table 1. The etymology for the species name T. taitungiacus comes from the region Taitung in Taiwan, where the organism was isolated from field soils (5). The recovery of this organism from Israel broadens the geographic region for this new species. T. taitungiacus has not been previously associated with human disease.

TABLE 1.

Comparison of features of thermophilic Thermoascus species

Feature	Characteristics of:
	T. taitungiacusa	T. crustaceusb	T. aegyptiacusc	T. aurantiacusd	T. thermophiluse
Conidiogenesis	Phialidic	Phialidic	Phialidic	Phialidic	Annellidic
Anamorph	P. taitungiacusf	P. crustaceus	P. aegyptiacusg	None or evanescent Paecilomyces-like	Polypaecilium species
9 days growth at:
16°C	−	−	−	−	−
20°C	−	+	−	−	±
25°C	+	+	+	−	+
30°C	+	+	+	−	+
35°C	+	+	+	+	+
40°C	+	+	+	+	+
50°C	+	+	+	+	+
LMh ascospore texture	Predominately echinulatei	Finely echinulate	Nearly smooth	Finely echinulate	Smooth
SEM ascospore texture	Irregularly verrucose	Finely echinulate	Slightly verruculose	Very minutely verrucosej	Verruculosek

^aT. taitungiacus K. Y. Chen and Z. C. Chen sp. nov. 

^bT. crustaceus (Apinis and Chesters) Stolk. 

^cT. aegyptiacus Ueda and Udagawa. 

^dT. aurantiacus Miehe. 

^eT. thermophilus (Sopp) von Arx. 

^fConidia of all Paecilomyces anamorphs were initially cylindrical, later becoming elliptical to subglobose or globose at maturity. 

^gNot all authorities believe this is a distinct species (13). 

^hLM, light microscopy. 

ⁱEchinulate is defined as having sharply pointed spines, delicately spiny. 

^jVerrucose indicates delicately verrucose. 

^kVerruculose is defined as having small, rounded processes or warts. 

Laboratory contamination is unlikely in the case presented, as T. taitungiacus was isolated in pure growth from four different dialysate fluids over a period of 8 days. Moreover, our patient had clinical and laboratory evidence of CPD-associated peritonitis and considerable morbidity due to a protracted symptomatic disease course and the development of a hypercatabolic state. In addition to Tenckhoff catheter removal, he required an extended period of systemic antifungal therapy and parenteral alimentation. Paecilomyces and Thermoascus species are a rare cause of fungal peritonitis (E. Bibashi, L. Sigler, E. Mitsopoulos, E. Roilides, M. Rinaldi, D. Sutton, D. Tsakiris, and M. Papadimitriou, Abstr. 13th Congr. Int. Soc. Hum. Anim. Mycol., abstr. P142, 1997). The first case was reported in 1990 by Lye (16). Since then only 14 additional cases have been published, 13 of which were identified as P. variotii (1, 3, 6, 8, 12, 15–18, 22). Paecilomyces peritonitis is associated with substantial morbidity. Ten of 14 reported patients had to be removed from CPD and were placed on hemodialysis (1, 6, 8, 12, 16–18); however, no fatalities were associated with Paecilomyces peritonitis. This stands in contrast with the high mortality rate (up to one-third of cases) reported in CPD-related peritonitis due to other fungi (4). The optimal approach for the treatment of Paecilomyces peritonitis is difficult to derive from the published data. Removal of the Tenckhoff catheter was the rule in the majority of the patients (1, 3, 6, 8, 12, 16–18); however, the antifungal regimens differed widely among the patients, even within the same center (17). Of note, one patient did not receive any antifungal therapy (18), while another received fluconazole despite the fact that his Paecilomyces isolate appeared resistant, in vitro, to the drug (8). All eight tested Paecilomyces isolates from the reported cases appeared susceptible, in vitro, to itraconazole (1, 3, 12, 17; Bibashi et al., Abstr. 13th Congr. Int. Soc. Hum. Anim. Mycol.). Eight of nine tested isolates were susceptible to amphotericin B (1, 8, 12, 17, 18, 22; Bibashi et al., Abstr. 13th Congr. Int. Soc. Hum. Anim. Mycol.), five of five to flucytosine (1, 3, 17), and only two of ten to fluconazole (1, 3, 8, 12, 17, 22; Bibashi et al., Abstr. 13th Congr. Int. Soc. Hum. Anim. Mycol.). T. taitungiacus can be added to an enlarging list of opportunistic filamentous fungi associated with human infection in general and CPD-associated peritonitis in particular. Further case studies are required to elucidate the clinical course of infection and the optimal treatment strategies.