Surg Today Jpn J Surg (1997) 27:816-825 SUI~ERvToI)AV

�9 Springer-Verlag 1997

Postoperative Enteritis Caused by Methicillin-Resistant Staphylococcus aureus

TAKASHI KODAMA 1, TAKAHIRO SANTO 1, TAKASHI YOKOYAMA 2, YOSHIO TAKESUE 1, EISO HIYAMA 2, YuJI IMAMURA 1, YOSHIAKI MURAKAMI 1, HIROAKI TSUMURA 1, KANAE SHINBARA 1, NAOKUNI TATSUMOTO 1, and YUICHIROU MATSUURA ~

~First Department of Surgery and 2Department of General Medicine, Hiroshima University School of Medicine, 1-2-3 Kasumi, Minami-ku, Hiroshima 734, Japan

Abstract: We examined the clinical features of 14 men (mean age 72 years) with postoperative enteritis caused by methicillin-resistant Staphylococcus aureus (MRSA). The patients had all undergone surgery for the treatment of diges- tive diseases and had received antibiotic prophylaxis consist- ing of an extended-spectrum cephem. Diarrhea appeared a mean of 3.3 days postoperatively and lasted for 5 days on average. In severe cases organ insufficiency was involved. Coagulase-positive staphylococci were the predominant or- ganisms isolated from watery diarrhea. In 13 of 14 patients, coagulase type II isolates producing enterotoxins A, C and toxic shock syndrome toxin-1 (TSST-1) with enterotoxin A, C, and tst genes were isolated. These strains were sensitive to vancomycin and arbekacin; however, they were highly resis- tant to many other antibiotics. We also investigated the effects of a glucocorticoid hormone and gamma globulin on pro- duction of tumor necrosis factor-or (TNF-e0 and interleukin -2 (IL-2) obtained from healthy volunteers. TNF-e~ and IL-2 production was enhanced by TSST-1 and the supernatant of Iscove-modified dulbecco medium, in which coagulase type II isolates producing enterotoxins A, C and TSST-1 with en- terotoxin A, C were cultured for 24h. Both glucocorticoid hormone and gamma globulin suppressed TNF-et and IL-2 production, thus suggesting that these drugs may be effective in treating postoperative MRSA enteritis.

Key Words: MRSA, cytokine, enterotoxin, glucocorticoide hormone, gamma globulin

Introduct ion

Coagulase-positive staphylococci resistant to many anti- biotics were isolated from many sources in the 1980s in Japan, and postoperative methicillin-resistant S t a p h y l o -

c o c c u s a u r e u s (MRSA) enteritis has been prevalent

Reprint requests to: T. Kodama (Received for publication on Mar. 25, 1996; accepted on Nov. 7, 1996)

since 1983. The clinical manifestations of MRSA enteritis range from mild disease to fatal illness. The reported mortality is approximately 10%. 1 How does MRSA enteritis develop? Toxic shock syndrome (TSS), a disease caused by S. a u r e u s , has been focused on as a key to solving this problem. Todd et al. have suggested that toxic shock syndrome (TSS) is associated with sta- phylococcal phage group I organisms, which can pro- duce potent new toxin(s) capable of causing severe multisystem injury. 2 Exotoxin, toxic shock syndrome toxin-1 (TSST-1), was found to be produced by S. a u r e u s isolated from TSS patients. Though many fea- tures of the disease remain unexplained, TSS is now believed to be toxin-mediated. 3,4 It has been reported that TSST-1 stimulates murine and human macro- phages to produce such monokines as interleukin-1 (IL-1) 5,6 and tumor necrosis factor-c~ (TNF-cQ 7 while it also stimulates murine and human T cells polyclonally to proliferate and to produce such lymphokines as interleukin-2 (IL-2) 8 and interferon-y. 9 It has been re- ported that lymphokines induce various adverse effects when administered to malignant tumor-bearing pa- tients, ml Many of the clinical signs and symptoms of those adverse effects are also observed in TSS patients. Based on these observations it is possible to hypothesize that lymphokines and monokines produced in amounts over the physiologically required level may play an important role in the development of the abnormal changes observed in TSS.

We previously demonstrated that there were various types of MRSA producing exotoxin, and MRSA that causes enteritis can produce TSST-1 and staphylococcal enterotoxin (SE) A and C. 12 Similarities between TSS and postoperative MRSA enteritis suggest that they may have a common pathology. Therefore, therapy di- rected toward toxin and cytokines may be a key in the management of postoperative MRSA enteritis. We hy- pothesize that IL-1, TNF-c~, and IL-2 might thus play a pathogenic role in MRSA enteritis. We examined the

T. Kodama et al.: Postoperative MRSA Enteritis 817

clinical manifestations of postoperative MRSA enteritis and investigated whether or not the supernatant of MRSA induces the production of IL-1, IL-2, and TNF- c~ in vitro, while also investigating the effects of gluco- corticoid hormone and gamma globulin on the release of IL-1, IL-2, and TNF-a in peripheral blood mono- nuclear cells (PBNMCs) stimulated by TSST-1 or the supernatant of the Iscove-modified Dulbecco medium, in which MRSA producing TSST-1, S E A and C were cultured under constant shaking for 24 h.

Patients and Methods

We examined the clinical features of 14 male patients who developed postoperative MRSA enteritis between 1 April 1990 and 31 December 1995. The patients ful- filled the following criteria for postoperative MRSA enteritis: (a) watery diarrhea and (b) dominant isolation of MRSA from stool specimens. The patients ranged from 57 to 88 years of age (mean age 72 years). They had all undergone surgery for the treating of digestive diseases and had received postoperative antibiotic pro- phylaxis consisting of a third-generation cephem. Can- cer of the stomach (8 patients) was the most common indication for surgery, and gastrectomy for cancer of the stomach was the most common surgical procedure.

Peripheral blood cell counts, multichannel blood chemistry tests, and a blood gas analysis were all per- formed. The severity of illness was determined based on our severity scoring system (Table 1). The progress of a given patient was traced by following the severity score, and repeated at 3- to 7-day intervals. Freshly passed diarrheal fecal material and blood were cultured ac- cording to standard techniques. Stock cultures were fro- zen in brain-heart infusion broth at -70~ Antibiotic sensitivities were evaluated by the standard method of the Japan Society of Chemotherapy. 13 The antibi- otics tested included methicillin (DMPPC), cloxacil- lin (MPIPC), cefazolin (CEZ), cefmenoxime (CMX), imipenem/cilastatin (IPM/CS), arbekacin (ABK), gen- tamicin (GM), amikacin (AMK), minocycline (MINO),

ofloxacin (OFLX), clindamycin (CLCM), and vancomy- cin (VCM).

Coagulase Typing of S. aureus

Clinical isolates were cultured under constant shaking for 24h in a heart infusion broth and centrifuged at 3 000 rpm for 30 min. Anticoagulase serum (type I-VIII) was added to the supernatant and the mixture was incu- bated at 37~ for i h. Coagulase typing of S. aureus was performed using the corresponding coagulase antise- rum that has been found to inhibit the coagulation of normal rabbit plasma.

SE and TSST-1 were grown with constant shaking for 24h in a heart infusion broth at 37~ SE was assayed in the supernatant fluid recovered after centrifugation (3 000rpm, 30min) by the reverse passive latex agglu- tination technique using polystyrene latex particles coupled with immunoglobulins of anti-SE A-D and TSST-1 rabbit hyperimmune serum fractionated by affinity chromatography.

The mec A gene, the SE gene, and the tst gene were analyzed by the polymerase chain reaction method.

Blood Collection and Mononuclear Cell Isolation and Stimulation

Normal PBNMCs were isolated from healthy human donors by Ficoll-Hypaque density sedimentation and cultured in flat-bottomed 16-well plates at 5 • 105 cells/ ml in Iscove-modified Dulbecco medium supplemented with 100 U/ml of penicillin and 100 ~g/ml of streptomy- cin for 24h with TSST-1 (100pg/ml to l~tg/ml) or the supernatant of MRSA (1 to 100 ~tl/ml) or 10~tl/ml of the supernatant of other-type MRSA in the presence or absence of methyl-prednisolone (0.25-2.5 mg/ml) or gamma globulin (0.625-6.25 mg/ml) (S-sulfated human immunoglobulin, Teijin, Tokyo, Japan). The superna- tant of the Iscove-modified Dulbecco medium, in which MRSA producing TSST-1, S E A and C were cultured under constant shaking for 24h, was defined as the supernatant of MRSA.

Table 1. Severity scores

Points

+4 +3 +2 +1 0 +1 +2 +3 +4

Oxygenation (Pao2, mmHg) Acid-base pH (blood) WBCs (xl03/mm 3) _->40 PLT (104/mm 3) Serum creatinine (mg/dl) ->_3.5 Total bilirubin (mg/dl) >=3.5 Blood sugar (mg/dl)

2.0-3.4 2.0-3.4

>70 61-70 55-60 <55 >_-7.6 7 .5-7.59 7.32-7.49 7.25-7.32 7.15-7.24 <7.15 20-39.9 15-19~9 3-14.9 1.0-2.9 <1

>=10 5.1-9.9 _--<5 1.5-1.9 _-<1.4 1.5-1.9 _-<1.4 __->200 151-199 __--<150

Paoz, arterial oxygen pressure; WBCs, white blood cells; PLT, platelets

818 T. Kodama et al.: Postoperative MRSA Enteritis

The supernatant of the Iscove-modified Dulbecco medium, in which MRSA (coagulase IV producing SE A, coagulase II producing SE B, coagulase II producing SE C and TSST-1, or coagulase VII) were cultured un- der constant shaking for 24h, was defined as the super- natant of other-type MRSA: all were isolated from many sources in our ward from 1984 to 1995. The cul- tures were incubated for 24h at 37~ in a humidified atmosphere containing 5% CO2 in air. The culture supernatants were collected, centrifuged at 800 • g for 7min, and kept at - 8 0 ~ until needed.

Assays

TNF-a was assayed by an enzyme-linked immuno- sorbent assay (ELISA) kit (Endogen, Boston, MA, USA) with reported lower limits of detection of 5- 10pg/ml. IL-10 was also assayed by the Endogen ELISA kit, and IL-2 was assayed by the Quantikine ELISA kit (Minneapolis, MN, USA).

Statistical Method

For the statistical analysis, the unpaired Student 's t-test was used and the values are presented as the mean + SE. The values were considered significant if P =< 0.05.

Results

Clinical Manifestations

All patients presented with acute onset of fever, ab- dominal cramps and distention, and watery diarrhea. The median interval between surgery and the onset of disease was 2 days. Diarrhea appeared a mean of 3.3 days postoperatively and lasted for 5 days on average. It was usually uncontrollable, often constant, and con- sisted of watery, brownish or green stool with relatively little odor. Patients with severe cases experienced as many as 20 bouts of watery diarrhea in a single day. In some patients, an excessive amount of fluid had been

(points) (points)

10 3"

8

2" �9 ~ 6 s

-,~

0 . . . . . 7.. 0 0 10 2 0 30 0 lO 20 30

Time(days) a b Time(days)

Fig. la,b. Changes in the severity score in patients with methicillin- resistant Staphylococcus aureus (MRSA) enteritis, a Changes in the severity score in 14 patients with MRSA enteritis, b Changes in the mean severity score in patients with MRSA enteritis (open circles) (n ~- 14) and in patients with abdominal sepsis (closed circles) (n = 10)

T a b l e 2. Laboratory findings

Patients with Test abnormal values Normal values

Patient values

Mean Range

Leukocytes 8/14 4900-9000 (/mm 3) Platelets 8/14 158--358 (xl03/mm 3) Total bilirubin 6/14 0.5-1.0 (mg/dl) Aspartate transaminase 11/14 8-40 (IU/1) Blood urea nitrogen 6/14 8-15 (mg/dl) Creatinine 4/14 0.57-0.87 (mg/dl) Blood sugar 9/11 70-120 (mg/dl) Pao z (room air) 11/12 90-100 (mmHg)

6329 20.1

1.7 100 17 0.79

184 69.3

2 900-16 900 53-510 0.5-5.3 12-523 7-31

0.5-1.57 100-322

51.2-86.2

T. Kodama et al.: Postoperative MRSA Enteritis 819

removed via gastric suction, following the onset of watery diarrhea. Hypotension and oliguria were present in 5 (35.7%) of 14 patients, and mild jaundice in 4 patients. None had any dermatologic manifestations such as rashes. Acquired respiratory distress syndrome (ARDS) developed in 6 patients, and 1 of these patients required ventilatory support. The patients received large amounts of intravenous fluids to maintain normal blood pressures and oral VCM to treat the enteritis. The mean severity score peaked on day 3 (Fig. 1). Patients

with postoperative MRSA enteritis had a high score on day 3 compared to patients with postoperative MRSA abdominal sepsis.

Laboratory Findings

At the onset of disease the leukocyte counts ranged from 2700 to 16900, with a marked nuclear left shift. The leukocyte counts were below 4000/mm 3 in 6 pa- tients and over 10000/mm 3 in only 2 patients. Platelet

t.

o.

100

80

60

40

20

0 0.1 0.2 0~9 0.78 1.56 3.1 6.25 12.5 25 50 100 100 �9

M i n i m u m inhibi tory concentrat ion ( m c g / m l )

tq

100

80

60

40

20

A

0.1 0.2 0.39 0.78 1.56 3.1 6.25 123 25 50 100 100 >

M i n i m u m inhibitory concentrat ion ( m c g / m l )

Fig. 2a,b. Minimal inhibitory concen- trations of coagulase-positive staphy- lococci associated with postoperative MRSA enteritis. Various drugs were tested against 14 isolates by the stan- dard method of the Japan Society of Chemotherapy. 17 a DMPPC, me- thicillin; MCIPC, cloxacillin; CEZ, cefazolin (closed squares); CMZ, cefmetazole (open squares); CMX, cefmenoxime (closed triangles); and IPM/CS, imipenem/cilastatin (open triangles), b GM, gentamicin (closed circles); ABK, arbekacin (open cir- cles); minocycline (closed squares); OFLX, ofloxacin (open squares); CLDM, clindamycin (closed tri- angles); and VCM, vancomycin (open triangles)

(pg/ml)

1200

T N F - A L P H A

(pg/ml)

1000

I L - 2

1000 800

800

600

a

600

400

200

0

c o n t r o l 0.1 1 1 0 1 0 0 1 0 0 0

Concentrat ion o f T S S T - l ( n g / m l )

4OO

200

c o n t r o l 0 . 1 1 1 0 1 0 0 1 0 0 0

Concentrat ion o f T S S T - I ( n g / m l )

Fig. 3a,b. Production of tumor ne- crosis factor alpha (TNF-a) and interleukin-2 (IL-2) by human pe- ripheral mononuclear cells in re- sponse to increasing concentrations of TSST-1. Human peripheral mononuclear cells (5 • 10S/ml) were cultured in fiat-bottomed 16-well plates with increasing concentrations of TSST-1 for 24 h and the production of TNF-c~ (a) and IL-2 (b) was determined by a cytokine-specific enzyme-linked immunoadsorbent assay (ELISA). Data are presented as mean _+ SE

820 T. Kodama et al.: Postoperative MRSA Enteritis

counts were below 100000/mm 3 in 3 patients. The total bilirubin level was mildly elevated in 4 patients. The serum creatinine level was not elevated. The arterial oxygen pressure (Pao2) was below 70mmHg (room air) in 4 patients (Table 2).

IL-2 decreased. Production of TNF-c~ and IL-2 by PBNMCs induced by the stimulation of 10gl/ml of supernatants (other types of MRSA) was lower than by PBNMCs induced by stimulation of 10~l/ml of supernatant (MRSA) (Fig. 5).

Bacteriological Aspects

Blood cultures were negative in all patients. S. aureus was isolated from stool specimens in all patients. All isolated strains of S. aureus were highly resistant (100 or >100~tg/ml) to IPM/CS, DMPPC, CEZ, CMZ, and CMX and tended to be resistant to CLDM, MINO, OFLX, GM, and AMK. S. aureus strains were sensitive to VCM and ABK (Fig. 2).

All isolates were coagulase type II. All strains except one produced SE A, C and TSST-1 with the enterotoxin- A,C and tst genes. One strain produced enterotixin C and TSST-1 with the enterotoxin C and tst genes. The mec A gene was detected in all strains.

Production of TNF-a, IL-lfl, and IL-2

TSST-1 stimulated TNF-a production by cultured PBNMCs in a concentration-dependent manner at doses ranging from 0.1 to 10ng/ml. Maximal TNF-et release in the ceils cultured for 24 h at 37~ occurred in response to TSST-1 concentrations of 10ng/ml, TSST-1 stimulation of IL-2 production by cultured PBNMCs was also concentration-dependent at doses ranging from 0.t to 100ng/ml (Fig. 3). Even 1 ~tl/ml of the super- natant (MRSA) significantly increased TNF-a, IL-l[3, and IL~2 production by PBNMCs. Maximal TNF-c~, IL- 1~, and IL-2 production was observed at a concentra- tion of 10~tl/ml (Fig. 4). At a supernatant (MRSA) concentration of 100F1/ml, production of TNF-ct or

Suppression o f lL-lfl, TNF-a, and IL-2 Production in PBNMCs by Methylprednisolone and Gamma Globulin

Gamma globulin in doses ranging from 0.625 to 6.25 mg/ ml inhibited 50% of TSST-1 (10ng/ml)-stimulated pro- duction of TNF-a. IL-2 production stimulated by 10ng/ ml TSST-1 was inhibited by gamma globulin in a concentration-dependent manner up to 6.25mg/ml of gamma globulin (Fig. 6). Methylprednisolone com- pletely inhibited TSST-1 (10ng/ml)-stimulated produc- tion of TNF-et and IL-2 at doses ranging from 0.25 to 2.5 mg/ml (Fig. 7).

At doses ranging from 0.625 to 6.25mg/ml, gamma globulin inhibited 70% of TNF-a production by PBNMCs stimulated by 10Fl/ml of the supernatant (MRSA). IL-2 production stimulated by 10~tl/ml of su- pernatant (MRSA) was completely inhibited by gamma globulin at doses of 0.625 to 6.25mg/ml. Gamma glo- bulin dose-dependently inhibited 70%-100% of IL- 113 production by PBNMCs induced by 10~tl/ml of supernatant (MRSA) at doses of 0.625-6.25mg/ml (Fig. 8).

Methylprednisolone (0.25-2.5 mg/ml) completely inhibited TNF-a and IL-2 production by PBNMCs induced by 10 ~tl/ml of the supernatant (MRSA). Meth- ylprednisolone (0.25-2.5mg/ml) inhibited IL-I~ pro- duction by PBNMCs induced by 10 ~tl/ml of supernatant (MRSA) below the control value (Fig. 9).

(pg/ml) IL-1 BETA

2000

1000

a - control IIAC 1 IIAC 10 IIAC 100 b Concentration of MRSA supernatant (micro I/ml)

(pg/ml) TNF-ALPHA (pg/ml) IL-2 1000, , 800

800

600

400 ]

200 j ~ .

o control IIAC1 IIAC1011AC100 Concentration of MRSA supernatant (micro I/ml)

Fig. 4a--e. Production of IL-I[3, TNF-ct, and IL-2 by human peripheral mononuclear cells in response to increasing con- centrations of the supernatant (MRSA). Human peripheral mononuclear cells (5 • 105/ml) were cultured in flat-bottomed 16-well plates with increasing concentrations of TSST-1 for 24h and the production of IL-113 (a), TNF-a (b), and IL-2

7001 600 5001 4001 300 200' 100"

control IIAC1 IIAC1011AC100 Concentration of MRSA supematant (micro I/ml)

(e) was determined by cytokine-specific ELISAs. The super- natant (MRSA) was obtained from the supernatant of the Iscove-modified Dulbecco medium, in which MRSAs produc- ing TSST-1, staphylococcal enterotoxin (SE) A and C were cultured under constant shaking for 24h. Data are presented as mean _+ SE

T. Kodama et al.: Postoperative MRSA Enteritis 821

(pg/ml) T N F

1000 i

(pg/ml)

800'3"

IL-2

800 700'

600'

400 300'

200 �84 200"

100'

0 ~ 0 ~ l , ~ i ~ n ~ a IIAC IVA IIB IIC VII control b I IAC IVA l iB IIC VII control

Fig. 5. Production of a TNF<t and b IL-2 by 10~d/ml of the supernatant (other-type MRSA)-activated human peripheral mononuclear cells. Human peripheral mononuclear cells (5 x 105/ml) were cultured in fiat-bottomed 16-well plates with 101xl/ml of the supernatant (other-type MRSA) for 24h and the production of TNF-e and IL-2 was determined by cytokine-specific ELISAs. The supernatant (other-type MRSA) was obtained from the supernatant of the Iscove- modified Dulbecco medium, in which coagulase IV type

MRSAs producing SE A, coagulase II type MRSAs producing SE B, coagulase II type MRSAs producing SE C, or coagulase VII type MRSAs were cultured under constant shaking for 24h. IIAC, coagulase II type MRSAs producing TSST-1, SE A and C;/VA, coagulase IV type MRSAs producing SE A; liB, coagulase II type MRSAs producing SE B; HC, coagulase II type MRSAs producing TSST-1, SE C; VII, coagulase VII type MRSAs producing non-SEs. Data are presented as mean -2-- SE

T NF

(pg/ml) (pg/ml) 1200 1000 ]

1000 800"

I L - 2

800

600

400

200

0 control TSST-1 0.625 1.25 6.25

gamma-globulin(mg/ml) +

a TSST-1

600"

400"

200"

controlTSST-1 0.625 1.25 6.25 gamma-globulin(mg/ml)

+ TSST-1

Fig. 6. Effects of gamma globulin on the production of a TNF-e and b IL-2 by TSST-activated mononu- clear cells. Human mononuclear cells (5 x 105) were activated by TSST-1 (100 ng/ml) and cultured with gamma globulin (0.625, 1.25, and 6.25 mg/ml) for 24h and the production of TNF-e and IL-2 in the supernatant was determined by cytokine-specific ELISAs. Data are presented as mean _+ SE

Discussion

Clinical Manifestations

Antibiotic-resistant Staphylococci enteritis has been observed in Japan since 1983.1 One of the most interest- ing aspects of this infection is that the majority of pa- tients who developed M R S A enteritis were men who

had received postoperat ive broad-spect rum antibiotic therapy. The barr ier effect against pathogens exerted by the normal stable flora appears to be extremely effec- tive against MRSA, and antibiotics effectively disrupt this barrier. Some broad-spect rum antibiotics that can alter the normal ecology of the bowel may permit an overgrowth of MRSA. It is important to differentiate be tween M R S A enteritis and pseudomembranous

T N F I L - 2

(pg/mi)

1200

1000

800

600

400

200

control TSST-1 0 . 2 5 0 .5 2 .5

melbylpr ed onisol one(mg]ml) +

TSST-I

(pg/ml)

1000

800

600

400

200

cont ro lTSST-I 0.25 0 .5 2.5 methylpredonisolone(mg/ml)

+

TSST-I

Fig. 7. Effects of methylpredniso- lone on the production of a TNF-ct and b IL-2 by TSST-activated mononuclear cells. Human mono- nuclear cells (5 x 105) were acti- vated by TSST-1 (100ng/ml) and cultured with methylprednisolone (0.25, 0.5, and 2.5mg/ml) for 24h and the production of TNF-ct and IL-2 in the supernatant was determined by cytokine-specific ELISAs. Data are presented as mean _+ SE

(pg/ml) IL-1 2500

2000

1500

1000

500

0 �84

a gamma-globulin(mg/ml) b +IIAC

(pg/ml) TNF 1000 T

800

600

400

200

0 . ~ . control IIAC 0.625 1.25 6.25 gamma-globulin(mg/ml)

+IIAC

Fig. 8. Effects of gamma globulin on the production of a IL- 113, b TNF-ct, and e IL-2 by supernatant (MRSA)-activated mononuclear cells. Human mononuclear cells (5 x 105) were activated by the supernatant (MRSA) (10 ~tg/ml) and cultured

(pg/ml) IL-2 800 700 600 500 400 300 200 100

0 control IIAC 0.625 1.25 6.25 gamma-globulin (mg/ml)

+IIAC

with gamma globulin (0.625, 1.25, and 6.5) for 24h and the production of IL-113, TNF-ct, and IL-2 in the supernatant was determined by cytokine-specific ELISAs. Data are presented as mean --- SE

(pg/ml) 2500

2000 t

1500 t

1000 t

500 "t --=

IL-1 (pg/ml)

1000

800

600

400

200

TNF (pg/ml) IL-2 800-

0 ~ a control IIAC 0.25 0.5 2.5 b

methylpredonisolone(mg/ml) +IIAC

700 600 500 400 300 2O0 100 S

control IIAC 0.25 0.5 2.5 control IIAC 0.25 0.5 2.5 C

methylpredonisolone(mg/ml) methylpredonisolone(mg/ml) +IIAC +IIAC

Fig. 9. Effects of methylprednisolone on the production of a IL-I~, b TNF-ct, and e IL-2 by supernatant (MRSA)-activated mononuclear cells. Human mononuclear cells (5 x 105) were activated by the supernatant (MRSA) (10 ~g/ml) and cultured

with methylprednisolone (0.25, 0.5, and 2.5mg/ml) for 24h and the production of IL-113, TNF-ct, and IL-2 in the super- natant was determined by cytokine-specific ELISAs. Data are presented as mean _+ SE

822

T. Kodama et al.: Postoperative MRSA Enteritis 823

colitis. In MRSA enteritis, the isolation rate of Clostridium difficile was very low and colonoscopy showed only edematous changes on the mucosa without any apparent pseudomembrane.

Postoperative MRSA enteritis is characterized by fe- ver, abdominal distention, and watery diarrhea, which appear on the 2nd or 3rd postoperative day, and subse- quently by oliguria, hypotension, hypoxia, and promi- nent leukopenia. If no appropriate therapy is instituted, ARDS, disseminating intravascular coagulation, and renal failure may develop. We evaluated the severity of illness using our scoring system, which is based on the derangement of the multiple organ system. A severity score of 4 or more was associated with a mortality rate of 50%. TM The severity score of MRSA enteritis on day 3 was 2.8 _ 2.3. The mean severity score of patients with other postoperative MRSA infections such as wound infection, abdominal sepsis, and anastomotic leakage, was lower than in patients with postoperative MRSA enteritis at our institution.

Biological Aspects

In the present study, MRSA was the predominant or- ganism isolated in cultures of freshly passed diarrheal fecal material. This isolated MRSA from enteritis was highly resistant to methicillin, cephems antibiotics, carbapenem, andtype II coagulase-producing SE A, C and TSST-1 with the enterotoxin-A, C and tst genes. The mec A gene was detected in all strains. The characteris- tics of the organisms that were isolated from MRSA enteritis were different from those of MRSA, which were coagulase IV type MRSAs producing SE A, co- agulase II type MRSAs producing SE B, coagulase II type MRSAs producing SE C, or coagulase VII type MRSAs, isolated from specimens of wound sepsis and abdominal sepsis.

These results thus led to the conclusion that MRSA enteritis is one of the nosocomial infections which can be caused by MRSA producing TSST-1, SE A and C and can thus even impair patients with a low severity score. Therefore, the management of nasocomial infec- tion, especially the restriction of prophylactic adminis- tration of antibiotics, is the most important factor for preventing MRSA enteritis.

Mechanism of Development of MRSA Enteritis

How does MRSA enteritis develop? TSS is being inves- tigated as a key to resolve this problem. TSS, which resembles MRSA enteritis in many features, is an acute and systemic illness caused by infection with a certain Staphylococcus strain. A toxin or toxins rather than simple bacteremia are involved as causative agents of this illness. An exotoxin, TSST-1, was also found to be

produced by S. aureus isolated from TSS patients. 3,4 Injected into rabbits, TSST-1 induces such symptoms as fever, diarrhea, low blood pressure, erythroderma, and respiratory distress. 15 Certain bacterial exotoxins such as TSST-1, a group of SE, are now designated bacterial super-antigens. 16 These exotoxins bind directly to MHC class II molecules and virtually activate full sets of T cells bearing particular V[3 elements. 16 A very small quantity of TSST-1 can stimulate macrophages and lym- phocytes to produce such cytokines as IL-1, TNF, and IL-2. 5~,1637 It has been reported that recombinant hu- man IL-2 or INF has various dose-related adverse ef- fects such as chills, high fever, malaise, headache, nausea, and vomiting, diarrhea, weight gain due to fluid retention, hypotension, and dyspnea when administered to patients with a variety of malignancies.m1 Many of the cl~ical signs and symptoms observed in those ad- verse effects are also observed in TSS patients. Al- though cytokines play a central role in the host's response to infection, lymphokines and monokines such as IL-2, IL-1, and TNF produced in amounts over the physiological required level have been considered to be involved in the mechanisms leading to the pathological changes observed in TSS. High levels of circulating cytokines, which can lead to subsequent organ dysfunc- tion, have been observed in patients with postoperative systemic inflammatory response syndrome (SIRS) or septic shock as well as TSS. 18-21 Though there is no proof that the mechanism of the severity of illness in MRSA enteritis is associated with toxins, this observation, in which the causative organisms produce TSST-1, S E A and C, thus suggests that toxins may play an important role in the development of MRSA enteritis as well as TSS. Therefore, it seems necessary to investigate the mode of action of TSST-1 and the supernatant of MRSA on IL-1, TNF, and IL-2 production by PBMNCs, as a step to understand the etiology of MRSA enteritis and also to reach a better understand- ing of the mechanisms involved in the response to toxins which would thus allow us to test the efficacy of thera- peutic modalities.

In Vitro Study of Cytokine Production by Stimulated PBMNCs

We determined that TSST-1, S E A and C was found to be produced by MRSA isolated from enteritis. Very small quantities of S E A and C as well as TSST-1 can stimulate macrophages and lymphocytes to produce such cytokines as IL-1, TNF, and IL-2. 5-8,16,17 In the present study, we demonstrated that the production of TNF-ct and IL-2 by PBNMCs was induced by TSST-1 and 10~d/ml of the supernatant (MRSA). Very small quantities of the supernatant (MRSA) as well as TSST- 1 can thus stimulate macrophages and lymphocytes to

824 T. Kodama et al.: Postoperat ive M R S A Enteri t is

produce such cytokines as IL-1, TNF, and IL-2. These findings thus suggest that toxins produced by MRSA may play an important role in the development of MRSA enteritis as well as TSS, and it is important to take the necessary steps to control any excessive cytokines if MRSA enteritis occurs. The production of TNF-ct and IL-2 by PBNMCs by 100~d/ml supernatant (MRSA) was less than that induced by 10 ~tl/ml of su- pernatant (MRSA). We thus speculate that cell death induced by the cytotoxic effect of the supernatant (MRSA), which contains leukolysin, may thus have in- hibited the production of TNF-a and IL-2 by PBNMCs. The production of TNF-a and IL-2 by PBNMCs induced by 10~tl/ml of the supernatants (other-type MRSA) was lower than that induced by 10~tl/ml of the supernatant (MRSA). Patients with other-type MRSA infections did not clinically become severe. We thus speculate that the severity of illness in postoperative infections with MRSA is related to the amount of cytokine production by toxins.

It has reported that in coincubation with TSST-1, hydrocortisone completely eliminated the enhancing effect of the monocyte supernatant on thymocyte prolif- eration. Host factors may thus potentiate the induction of IL-1 by TSST-1 in vivo, and therefore the serum antibody to TSST-1 may have a protective effect. 6 We determined the effects of gamma globulin and me- thylprednisolone on the production of cytokines by stimulated PBMNCs. We found that the production of TNF-a and IL-2 by PBNMCs induced by TSST-1 or the supernatant (MRSA) was inhibited by gamma globu- lin. These findings suggest that the administration of gamma globulin may be effective in patients with post- operative MRSA enteritis.

We hypothesize that this effect was due to the into- xication of the MRSA toxin by gamma globulin, be- cause gamma globulin did not have the same effect in lipopolysaccharide (LPS)-stimulated PBNMCs. The titers of antibody to TSST-1 are high after both the intramuscular and intravenous administration of gamma globulin. 6 In addition, the antibody titers are 1 : 100 or higher in approximately 80% of units of frozen plasma; 79% of normal adults have been shown to have antibody titers of 1 : 100 or higher, whereas 83% of pa- tients with toxic shock syndrome have acute phase titers of l : 5 or less. 3

The production of cytokines in monocytes is sup- pressed by steroidsY In the present study, the produc- tion of TNF-a and IL-2 induced by TSST-1 or the supernatant (MRSA) was completely suppressed by methylprednisolone, thus suggesting that the adminis- tration of methylprednisolone may be effective in pa- tients with postoperative MRSA enteritis.

The use of steroids to treat infectious diseases is con- troversial. Our findings suggest that short-term steroid

therapy with oral VCM does not worsen postoperative MRSA enteritis and was even beneficial in patients with MRSA enteritis. In fact, the administration of gluco- corticoid hormone and gamma globulin was found to benefit patients with MRSA enteritis who developed ARDS and oliguria.

Finally, we speculate that toxins are initially pro- duced in the gastrointestinal tract and may then translo- care from the gut to the bloodstream, thus inducing various biological responses. We did not detect any cytokines in the toxic phase. There was also no evidence that toxins produce cytokines in vivo. It would have been of great interest to do a blood mediator analysis. Therefore, additional studies will be required to ascer- tain the precise manner in which the toxins produced by MRSA cause multiple systemic effects in humans.

References

1. Hori K, Yura J, Shinagawa N, Sakurai S, Mashita K, Mizuno A (1989) Postoperative enterocolitis and the current status of MRSA enteritis - - The results of a questionnaire survey in Japan. J Jpn A Infect Dis 63:701-707

2. Todd J, Fishaut M, Kapral F, Welch T (1978) Toxic-shock syn- drome associated with phage-group-1 staphlococci. Lancet 2: 1116-1118

3. Bergdoll MS, Crass BA, Reiser RF, Robbins RN, Davis JP (1981) A new staphylococcal enterotoxin, enterotoxin F, associated with toxic shock syndrome Staphylococcus aureus isolates. Lancet 1:1017-1021

4. Schlievert PM, Shands KN, Dan BB, Schmid GP, Nishimura RD (1981) Identification and characterization of an exotoxin from Staphylococcus aureus associated with toxic-shock syndrome. J Infect Dis 143:509-516

5. Ikejima T, Dinarello CA, Gill DM, Wolff SM (1984) Induction of human interleukin-1 by a product of Staphylococcus aureus associated with toxic shock syndrome. J Clin Invest 73:1312- 1320

6. Parsonnet J, Hickman RK, Eardley DD, Pier GB (1985) Induc- tion of human interleukin-1 by toxic-shock-syndrome toxin-1. J Infect Dis 151:514-522

7. Parsonnet J, Gillis ZA (1988) Production of tumor necrosis factor by human monocytes in response to toxic-shock-syndrome toxin- 1. J Infect Dis 158:1026-1033

8. Micusan VV, Mercier G, Bhatti AR, Reiser RF, Bergdoil MS, Oth D (1986) Production of human and murine interleukin-2 by toxic shock syndrome toxin-1. Immunology 58:203-208

9. Jupin BC, Anderson S, Damais C, Alouf JE, Parant M (1988) Toxic shock syndrome toxin 1 as an inducer of human tumor necrosis factors and ),-interferon. J Exp Med 167:752-761

10. Gutterman JU, Blumenschein GR, Alexanian R, Yap HY, Buzdar AU, Cabanillas F, Hortobaggi GN, Hersh EM, Rasmussen SL, Harmon M, Kramer M, Pestka S (1980) Leuko- cyte interferon-induced tumor regression in human metastatic breast cancer, multiple myeloma and malignant lymphoma. Ann Int Med 93:399-406

11. Lotze MT, Matory YL, Ettinghausen SE, Rayner AA, Sharrow SO, Seipp CAY, Custer MC, Rosenberg SA (1985) In vivo admin- istration of purified human interleukin-2. II. Half life, immuno- logic effect, and expansion of peripheral lymphoid cells in vivo with recombinant IL-2. J Immunol 135:2865-2875

12. Takesue Y, Yokoyama T, Kodama T, Santou T, Nakamitsu A, Murakami Y, Immamura Y, Miyamoto K, Okita M, Tsumura H,

T. Kodama et al.: Postoperative MRSA Enteritis 825

Itaha H, Matsuura Y (1991) Toxin involvement in methicillin- resistant Staphylococcus aureus enteritis in gastroenterological surgery. Gastroenterol Jpn 26:716-720

13. Fujii Y (1981) On the renewal of the standard method for MIC measurement. Chemotherapy 29:76-79

14. Kodama T, Yokoyama T, Takesue Y, Okita M, Nakamitsu A, Hiyama E, Immamura Y, Santou T, Murakarni Y, Tsumura H, Shinbara K, Miyamoto K, Tatsumoto N, Matsuura Y (1995) Se- verity and predicted outcome of postoperative Pseudomonas aeruginosa infections. Surg Today 25:672-678

15. Igarashi H, Fujikawa H, Usami H, Kawabata S, Morita T (1984) Purification and characterization of Staphylococcus aureus FRI 1169 and 587 toxic shock syndrome exotoxins. Infect Immun 44:175-181

16. Yan XJ, Li XY, Imanishi K, Kumazawa Y, Uchiyama T (1993) Study of activation of murine T cells with bacterial superantigen. J Immunol 150:3873-3881

17. Marrack P, Kappler J (1990) The staphylococcal enterotoxins and their relatives. Science 705-711

18. Mathison JC, Wolfson E, Ulevitch (1988) Participation of tumor necrosis factor in the mediation of Gram negative bacterial lipopolysaccharide-induced injury in rabbits. J Clin Invest 81:1925-1937

19. Hesse DG, Tracey KJ, Fong Y, Manogue KR, Palladino MA, Cerami A, Shires GT, Lowry SF (1988) Cytokine appearance in human endotoxemia and primate bacteremia. Surg Gynecol Obstet 166:147-153

20. Welbourn R, Goldman G, Kobzik L, Paterson I, Shepro D, Hechtman HB (1991) Interleukin-2 induces early multisystem organ edema mediated by neutrophils. Ann Surg 214:181-185

21. Ertel W, Morrison MH, Wang P, Zheng FBA, Ayala A, Chaudry IC (1991) The complex pattern of cytokines in sepsis. Association between prostaglandins, cachectin, and interleukins. Ann Surg 214:14-17

22. Lew W, Oppenheim JJ, Matsushima K (1988) Analysis of the suppression of IL-lct and IL-I~ production in human peripheral blood mononuclear adherent cells by a glucocorticoid hormone. J Immunol 140:1895-1902