点击显示 收起
Centre for Emerging Infectious Diseases, Department of Microbiology, School of Public Health
Clinical Immunology Unit, The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, New Territories, Hong Kong Special Administrative Region, People's Republic of China
The maturation of virus-specific immunoglobulin G avidity during severe acute respiratory syndromeassociated coronavirus infection was examined. The avidity indices were low (mean ± SD, 30.8% ± 11.6%) among serum samples collected 50 days after fever onset, intermediate (mean ± SD, 52.1% ± 14.1%) among samples collected between days 51 and 90, and high (mean ± SD, 78.1% ± 8.0%) among samples collected after day 90. Avidity indices of 40% and 55% could be considered as cutoff values for determination of recent (50 days) and past (>65 days) infection, respectively. Measurement of antibody avidity can be used to differentiate primary infection from reexposure and to assess humoral responses to candidate vaccines.
Since the worldwide outbreak of severe acute respiratory syndrome (SARS) between November 2002 and June 2003 [1], subsequent smaller outbreaks have occurred as a result of laboratory negligence or reemergence of SARS-associated coronavirus (SARS-CoV) from the natural reservoir [25]. These incidences bear witness to the fact that reemergence of SARS-CoV infection in humans is a real concern. Experience from the Guandong outbreak (which occurred between December 2003 and January 2004) suggests that the clinical presentation of disease and the transmission behavior of the reemerged SARS-CoV strain can be different from what was known before [4]. When a SARS outbreak occurs again, it is mandatory that a serological survey be conducted, to define the epidemiological character of the outbreak. Since these outbreaks may happen in places where a proportion of the population was exposed to the virus during a previous outbreak of SARS, a reliable method for differentiating between recent infection and past exposure is vital if a meaningful interpretation is to result from such investigations [4]. The avidity (functional affinity) of an antibody is a measure of the overall strength of interaction between antibody and antigen. The avidity of virus-specific IgG antibody is low during primary viral infection and increases with time [68]. However, exceptions to this rule have been observed for some viruses [9, 10]. Here, we report the maturation pattern of antiSARS-CoV nucleocapsid proteinspecific IgG antibody (hereafter, "antiSARS-CoV IgG antibody") avidity over the course of a 10-month period after primary infection and discuss the potential applications of our findings.
Patients, materials, and methods.
Sixty-one patients with SARS were recruited into the present study; they ranged in age from 21 to 81 years (mean ± SD, 37.0 ± 14.9 years), and 54.1% were female. These patients presented with acute-onset fever that progressed to pneumonia, which was otherwise unexplained. Nine patients (14.8%) required intensive care, all of whom eventually recovered. All 61 patients fulfilled the US Centers for Disease Control and Prevention criteria for SARS [11] and had serological evidence of SARS-CoV infection, as determined by the antiSARS-CoV IgG antibody immunofluorescence assay described elsewhere [12]. Forty-one patients (67.2%) seroconverted, and 20 (32.8%) developed a 4-fold increase in antibody titer. A total of 90 serum samples were available from the 61 patients, of whom 26 provided >1 sample for the study.
AntiSARS-CoV IgG antibody avidity was measured by a 2-step approach. The first step was to assess the concentrations of antiSARS-CoV IgG antibody in samples, so that samples that required further dilution for testing during the second step could be identified. On the basis of our serial-dilution experiments, samples with ODs of >2.5 needed to be further diluted to provide a linear range for measurement of avidity during the second step. Concentrations of antiSARS-CoV IgG antibody were measured by a recombinant nucleocapsid proteinbased enzyme immunoassay, ELISARS (IgGENE), in accordance with the manufacturer's instructions. Briefly, samples were diluted to a concentration of 1 : 50 by mixing 22 L of serum with 1.1 mL of sample diluent. An 100-L aliquot of the prediluted serum was added to an antigen-coated well. After incubation at room temperature (25°C) for 30 min, the wells were washed 3 times with the washing buffer provided. Antihuman IgG antibodies conjugated with horseradish peroxidase were added and were incubated at room temperature for 15 min. After a second washing step, 3,3,5,5-tetramethylbenzidine was added as a substrate for color development. Optical density was measured at 450 nm.
The second step was also based on the ELISARS assay but included a urea-elution procedure. Briefly, samples were further diluted, if necessary, according to the results obtained during the first step. The neat or prediluted serum samples were mixed with sample diluent as described above. The sample mixtures were added in duplicate to 2 antigen-coated wells. After the first incubation step, 300 L of urea was added to one of the wells, whereas the same volume of washing buffer was added to the other well, which served as a reference. On the basis of our initial optimization experiments using 5 early and 5 late samples (collected <20 and >250 days after fever onset, respectively), a soaking step at room temperature for 10 min with 4 mol/L urea diluted in washing buffer was found to be most suitable and, thus, was used in the present study. The urea-soaking step was followed by washing 3 times with the washing buffer provided. The subsequent conjugate-addition and color-development steps were conducted in accordance with the manufacturer's protocol. The antibody avidity index was calculated as ODurea/ODreference and is expressed as a percentage.
Samples collected 50 days after fever onset were also tested for antiSARS-CoV IgM antibody, so that IgM antibody detection and IgG antibody avidity measurement could be compared with respect to demonstrating a recent infection. AntiSARS-CoV IgM antibody was also detected by the ELISARS assay. Briefly, samples were treated by use of a rheumatoid factor removal kit (Chemicon) and then mixed with sample diluent to a final concentration of 1 : 50. A 100-L aliquot of the diluted sample was added to an antigen-coated well and subjected to the incubation, wash, and color-development steps described above, except that antihuman IgM antibody was used as the conjugate.
AntiSARS-CoV IgG antibody titers for paired serum samples were measured by an in-house indirect immunofluorescence assay that has been described elsewhere [12]. This was done so that the value of using changes in IgG antibody titers and that in antibody avidity could be compared with respect to demonstrating a recent infection.
Results.
For the 90 serum samples, the optical-density values obtained during the first step ranged from 0.639 to 3.509 (mean ± SD, 2.057 ± 0.862); 31 samples had an OD of >2.5 and, thus, required further dilution for the second step. Of these, 19 required further dilution of 1 : 2, and 12 required further dilution of 1 : 5, to achieve a reference OD of 2.5 before they were subjected to the second step for measurement of antibody avidity.
Of the 45 samples collected 50 days after fever onset, only 18 (40.0%) were positive for antiSARS-CoV IgM antibody, as determined by the ELISARS assay. For the 18 IgM-positive samples, the avidity indices ranged from 13.3% to 46.8%, with a median of 32.3% and an interquartile range of 23.1%37.6%. For the 27 IgM-negative samples, the avidity indices ranged from 7.4% to 51.8%, with a median of 31.9% and an interquartile range of 21.8%40.8%. There was no significant difference in avidity level between these 2 groups of samples (P = .746, Mann-Whitney U test).
Discussion.
Our data show that antiSARS-CoV IgG antibody avidity is low during primary infection and increases with time in a unidirectional manner. On the basis of this phenomenon, measurement of antibody avidity can be used to resolve certain difficulties that may be encountered in assessment of SARS-CoV infection. First, it can be used to differentiate between primary infection and reexposure. Although it was not possible to include patients who had been reexposed in the present study, on the basis of experience with other viral infections that have a similar pattern of antibody avidity maturation [13], it is reasonable to infer that patients reexposed to SARS-CoV will mount a humoral memory immune response that includes the production of antibodies with high avidity within a short period of time. Second, the presence of antibodies with low avidity could provide alternative evidence for demonstrating a primary infection when the IgM assay result is in doubt. This is important, given that viral serological testing based solely on the determination of the presence of IgM can lead to false conclusions, because IgM responses last for only a very short period of time and could be missed if serum samples are collected too early or too late [14]. On the other hand, IgM can persist for months or even years after primary infection and reappear during secondary infection [15]. When interpreting our IgM results, one should be aware that the IgM assay used in the present study was based on the indirect enzyme immunoassay format, the sensitivity of which might be inferior to that of the IgM capture format, and that our omission of the IgG antibody removal step might have decreased the assay's sensitivity. Nevertheless, our IgM assay results for the 45 samples collected 50 days after fever onset support the view that low antibody avidity could be a valuable alternative marker for defining primary infection, in particular when serum sample availability is limited in terms of collection time points. Our data show that all 36 samples with an avidity index of <40% were collected before day 50, whereas all 39 samples with an avidity index of >55% were collected after day 65. Avidity indices between 40% and 55% could be considered to represent "the maturation zone," in which the correlation between avidity and time since infection is less strong. Third, our comparison of the avidity indices for the 26 paired samples indicated that this approach could provide a helpful alternative to the use of increasing antibody concentration as serological evidence of recent infection. This is particularly important if convalescent samples are collected when antibody concentrations are no longer increasing, as was the case for most of our paired samples. Other possible applications of measurement of antibody avidity to SARS-CoV infection include use of the technique to assess humoral responses to vaccine candidates and to discriminate between primary and secondary vaccine failures.
References
1. World Health Organization. Summary of probable SARS cases with onset of illness from 1 November 2002 to 31 July 2003. Available at: http://www.who.int/csr/sars/country/table2003_09_23/en/print.html. Accessed 30 December 2003. First citation in article
2. Lim PL, Kurup A, Gopalakrishna G, et al. Laboratory-acquired severe acute respiratory syndrome. N Engl J Med 2004; 350:17405. First citation in article
3. World Health Organization. Severe acute respiratory syndrome (SARS) in Taiwan, China. Available at: http://www.who.int/csr/don/2003_12_17/en/. Accessed 5 August 2004. First citation in article
4. World Health Organization. Update 4: review of probable and laboratory-confirmed SARS cases in southern China. Available at: http://www.who.int/csr/don/2004_01_27/en/. Accessed 5 August 2004. First citation in article
5. World Health Organization. Investigation into China's recent SARS outbreak yields important lessons for global public health. Available at: http://www.wpro.who.int/sars/docs/update/update_07022004.asp. Accessed 5 August 2004. First citation in article
6. Inoue S, Hasegawa A, Matsuno S, Katow S. Changes in antibody avidity after virus infections: detection by an immunosorbent assay in which a mild protein-denaturing agent is employed. J Clin Microbiol 1984; 20:5259. First citation in article
7. Hedman K, Lappalaninen M, Soderlund M, Hedman L. Avidity of IgG in serodiagnosis of infectious diseases. Rev Med Microbiol 1993; 4:1239. First citation in article
8. Schubert J, Zens W, Weissbrich B. Comparative evaluation of the use of immunoblots and of IgG avidity assays as confirmatory tests for the diagnosis of acute EBV infections. J Clin Virol 1998; 11:16172. First citation in article
9. Thomas HI, Wilson S, O'Toole CM, et al. Differential maturation of avidity of IgG antibodies to gp41, p24 and p17 following infection with HIV-1. Clin Exp Immunol 1996; 103:18591. First citation in article
10. Thomas HI. The relative functional affinity of specific anti-core IgG in different categories of hepatitis B virus infection. J Med Virol 1997; 51:18997. First citation in article
11. Centers for Disease Control and Prevention. Severe acute respiratory syndrome. Available at: http://www.cdc.gov/ncidod/sars/guidance/B/pdf/app1.pdf. Accessed 3 March 2004. First citation in article
12. Chan PK, Ng KC, Chan RC, et al. Immunofluorescence assay for serologic diagnosis of SARS. Emerg Infect Dis 2004; 10:5302. First citation in article
13. Hashido M, Inouye S, Kawana T. Differentiation of primary from nonprimary genital herpes infections by a herpes simplex virus-specific immunoglobulin G avidity assay. J Clin Microbiol 1997; 35:17668. First citation in article
14. Brackmann HH, Oldenburg J, Eis-Hubinger AM, Gerritzen A, Hammerstein U, Hanfland P. Hepatitis A virus infection among the hemophilia population at the Bonn Hemophilia Center. Vox Sang 1994; 67(Suppl 1):S37. First citation in article
15. Roggendrof M, Heinz F, Deinhardt F, Kunz C. Serological diagnosis of tick-borne encephalitis by determination of antibodies of the IgM class. J Med Virol 1981; 7:4150. First citation in article
Saline-Filled Breast Implant Contamination with Curvularia Species among Women Who Underwent Cosmetic Breast Augmentation
Marion A. Kainer,1,2,5,a Homa Keshavarz,1,4 Bette J. Jensen,2 Matthew J. Arduino,2 Mary E. Brandt,3 Arvind A. Padhye,3 William R. Jarvis,2,b and Lennox K. Archibald2,a
1Epidemic Intelligence Service, Division of Applied Public Health Training, Epidemiology Program Office, and Divisions of 2Healthcare Quality Promotion, 3Bacterial and Mycotic Diseases, and 4Public Health Surveillance and Informatics, Epidemiology Program Office, Centers for Disease Control and Prevention, Atlanta, Georgia; 5Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Australia
Background.
During December 2000July 2001, black sediment was noted in saline-filled silicone breast implants of women who had undergone revision surgery at facility A. Curvularia fungus was isolated from implant saline.
Methods.
To identify risk factors for contamination with Curvularia species, we performed case-control, retrospective cohort, and laboratory studies and conducted procedural reviews. A case patient was defined as any woman who underwent revision surgery at facility A between January 2000 and June 2001 and had black sediment in her implants.
Results.
Five patients met the case definition. Contamination was associated with having had surgery performed in operating room (OR) 2 (4/88 vs. 1/140; P = .07) and a longer duration of surgery (P < .001). A longer duration spent in the OR was an additional risk factor (P = .005). Curvularia fungus was isolated from the sterile supply room, where saline bottles had been stored under a water-damaged ceiling, and from the corridor outside OR 2; it was also found more commonly from facility A personnel than from nonfacility A personnel (12/34 vs. 4/60; P < .001). Saline was warmed in a cabinet opposite OR 2, which was maintained at negative pressure differentials, then was poured into bowls open to the OR 2 environment before injection into implants.
Conclusion.
Surgeons should always use closed systems to inflate breast implants. Surgery center infection control measures must include moisture control and balanced ventilation systems.
--------------------------------------------------------------------------------
Received 20 September 2004; accepted 28 January 2005; electronically published 18 May 2005.
Presented in part: 40th Annual Meeting of the Infectious Diseases Society of America, Chicago, 2427 October 2002 (abstract 416).
Financial support: The investigation upon which this study is based was conducted while the authors were employees of the Centers for Disease Control and Prevention. The US Public Health Service underwrote all funding for this investigation.
Since submitting this manuscript, M.A.K. has provided professional clarification and interpretation on the results of this investigation to attorneys involved in pending court cases regarding this incident.
The use of trade names is for identification only and does not imply endorsement by the US Public Health Service or the US Department of Health and Human Services.
a Present affiliations: Tennessee Department of Health, Nashville, Tennessee (M.A.K.); Regeneration Technologies, Inc., Alachua, Florida (L.K.A.).
b W.R.J. is now retired and in private practice.
Reprints or correspondence: Dr. Marion A. Kainer, Tennessee Dept. of Health, 4th Fl., Cordell Hull Bldg., 425 5th Ave. N, Nashville, TN 37247 (marion.kainer@state.tn.us).
--------------------------------------------------------------------------------
During 2000, 187,755 women underwent cosmetic breast augmentation (CBA) surgery in the United States, an increase of 476% compared with 1992 [1, 2]. Most augmentations (94%) used saline-filled silicone shell prosthetic breast implants (SALBIs) [1]. Recognized complications with SALBIs include leakage, capsular contracture, and infection [3]. The incidence of infection after CBA surgery is <2% [47], and there have been few case reports of fungal contamination of SALBIs [810]. In this article, we describe an investigation of an outbreak of Curvularia species contamination of SALBIs of women who had undergone CBA surgery.
From 22 December 2000 through 21 July 2001, macroscopically visible black sediment was noted in the SALBIs in 5 women who had undergone revision surgery and SALBI removal. Curvularia fungus, a fast-growing dematiaceous fungus commonly found in soil, was cultured from the implant fluid in all 5 women. Two surgeons (surgeons X and Y) performed CBA procedures at a freestanding ambulatory surgical facility (facility A) and at medical center B, located 2 miles away. Surgeon X had performed all 5 initial CBAs at facility A during 1 June12 September 2000. When preliminary investigations by facility A personnel did not identify a source of contamination, the Centers for Disease Control and Prevention (CDC) was invited to assist in an investigation. We report the findings of the outbreak investigation conducted in JulyAugust 2001, which identified the source and risk factors for contamination with Curvularia species, and provide recommendations for the prevention of fungal contamination of operating rooms and SALBIs and future similar outbreaks.
PATIENTS AND METHODS
To identify risk factors for contamination of SALBIs with Curvularia species, we conducted analytic epidemiologic (case-control and retrospective cohort) studies, review of practices and procedures at facility A, and laboratory studies.
Case Definition and Ascertainment
A case patient was defined as any woman with macroscopic fungal contamination of 1 of her SALBIs who previously underwent CBA surgery at facility A between 1 January 2000 and 28 June 2001. To ascertain case patients, we interviewed surgical and operating room (OR) staff and reviewed microbiology laboratory and patient medical records.
Epidemiologic Studies
Case-control study.
A control patient was defined as a woman with 2 macroscopically clear SALBIs, as determined by direct visual inspection during revision surgery at facility A or medical center B. Both case and control patients had to have had a revision of a previous CBA procedure. Factors assessed included patient demographic data; indication for surgery; date, time, and duration of the surgical procedure; incision site; type of SALBI (manufacturer, lot number, style, and size); volume of saline injected; OR; operating staff; ancillary OR staff; and severity of illness, as measured by the American Society of Anesthesiologists (ASA) score. "OR time" was defined as the interval between arrival in the OR and incision closure.
Cohort study.
Because the risk for fungal contamination might have been present only during MaySeptember 2000 (i.e., the cohort study period), we conducted a cohort study of women whose original CBA surgery was performed by surgeon X during this period. Women who had had SALBIs but did not undergo revision surgery and remained asymptomatic were deemed noncase patients. We recognized a priori that misclassifying case patients as presumed noncase patients would bias our findings toward the null hypothesis that there was no difference between case and noncase patients. Because of this limitation, we also considered the direction and magnitude of statistical associations.
Review of Practices and Procedures
Surgical procedures and practice review.
We interviewed the surgeons and OR ancillary staff and observed a CBA procedure performed by surgeon X. We reviewed facility A's architectural plans and inspected the entire facility. We traced and recorded the path through which saline bottles and packaged SALBIs were transferred from storage in the sterile supply room to the OR.
Health care worker assessment and survey.
We administered a questionnaire to facility A clinical and nonclinical staff. Information collected included history of sinus or skin problems, fungal infections, use of artificial fingernails, and exposure to plants or soil. We also examined the hands and fingernails of facility A staff.
Facility A Review
We reviewed the layout of the ORs at facility A. In addition, we inspected the heating, ventilation, and air conditioning (HVAC) systems and the air duct interiors on 6 August 2001. We reviewed available engineering, general, and preventive maintenance records and specifications of the HVAC unit at facility A. Flutter strips were used to verify the direction of airflow from ORs to the corridor.
Microbiologic Studies
Personnel cultures.
In August 2001, we cultured nasal mucus, fingernail clippings, and hands (by the Handiwipe method [11]) of facility A personnel. To determine the prevalence of nasal colonization by Curvularia species in this region of the United States, we obtained cultures from OR personnel at medical center B, none of whom had worked at facility A during the preceding 12 months.
To establish whether colonized facility A staff could potentially shed Curvularia fungus, we placed settle plates at operating-table height in a small, designated room at another facility and left them for 20 min with the room empty. Next, the procedure was repeated with a person in the room; new settle plates were placed in the same format and for the same duration, within a 3-foot radius of the person being assessed.
Environmental cultures.
In August 2001, we obtained targeted surface cultures of the sterile supply room, saline bottles, a saline bottle warmer, inlet and outlet ventilation grilles, air duct interiors, and moist or water-damaged walls and ceilings. We cultured water condensate specimens from OR portable dehumidifiers. Settle plates were placed in the sterile supply room and the saline warmer for 20 min. Air sampling was performed using an air sampler (SAS Super 90; International PBI).
Statistical Analysis
Data were collected on standardized forms and analyzed using Epi Info (version 6.04; CDC) and SAS statistical software (version 6.12; SAS Institute). The 2 test or Fisher's exact test, where appropriate, was used to compare categorical variables; the Kruskal-Wallis test was used to compare continuous variables. Because fungal contamination involving 2 SALBIs in the same patient were not independent events, analyses were performed at the patient level and not at the breast implant level.
RESULTS
Case Patient Characteristics
Five patients met the case definition. Two were symptomatic, and 3 were asymptomaticthat is, SALBI fungal contamination was an incidental finding at revision surgery (table 1). The median age of case patients was 24 years (range, 2149 years). The 5 explanted SALBIs were round and smooth; 3 were manufactured by the Mentor Corporation and 2 were manufactured by the McGhan Medical Corporation; the lot numbers for implants of each brand differed from one another. Normal sterile saline for irrigation (Baxter Healthcare Corporation) was used to fill the breast implants in the OR; lot and serial numbers were not available for the saline bottles.
One woman had her initial CBA surgery in 1995 and underwent her first surgical revision in June 2000, to correct symmastia (i.e., medial confluence of the breasts producing a web across the midline). At her revision surgery in June 2000, both SALBIs were macroscopically clear. Saline was removed from each SALBI and replaced with fresh normal saline; the 2 SALBIs were then reimplanted. When this woman underwent further revision surgery, for cosmetic reasons, in July 2001, 1 of the 2 SALBIs contained black sediment that was macroscopically apparent and was later identified as Curvularia lunata (figure 1).
Epidemiologic Studies
Case-control study.
The 5 case patients and 17 control patients were similar in age, ASA score, volume of saline used to fill the breast implants, time of surgical procedure, and rank placement (e.g., first) on the operating list. In contrast, case patients were more likely than control patients to have undergone their operation on or after 1 May 2000 (5/5 vs. 5/17; odds ratio, undefined; P = .01) (figure 2), to have had their surgical procedure performed by surgeon X (5/5 vs. 4/17; odds ratio, undefined; P < .005), to have had a periareolar incision (4/5 vs. 2/17; odds ratio, 30.0 [95% confidence interval {CI}, 2.1421]; P < .01), and to have undergone 2 procedures at the same time as CBA (2/5 vs. 0/17; odds ratio, undefined; P < .05). In addition, case patients were significantly more likely than control patients to have had a longer duration of anesthesia (median, 82 vs. 62 min; P < .05) and duration of surgery (median, 66 vs. 45 min; P < .05) and to have a longer OR time (median, 107 vs. 67 min; P < .05).
Of the 5 case patients, 4 had surgery performed in OR 2, whereas 6 of 17 control patients had surgery performed in OR 2 (odds ratio, 7.3 [95% CI, 0.6681.4]; P = .14). Compared with patients operated on by surgeon Y, patients operated on by surgeon X had significantly longer OR (median, 107 vs. 62 min; P < .001), anesthesia (median, 85 vs. 61 min; P < .001), and surgery (median, 66 vs. 40 min; P < .001) times. In addition, longer OR times were significantly associated with additional procedures, such as mastopexy (median, 191 vs. 67 min; P < .05) and periareolar incisions (median, 159 vs. 69 min; P < .001). The periareolar approach was used only by surgeon X (6/9 vs. 0/13; P = .001). When the analysis was restricted to surgeon X, 4 of 5 case patients versus 1 of 4 control patients had surgery performed in OR 2 (odds ratio, 25 [95% CI, 0.5280.1]; P = .2). Indications for revision surgery for the 17 control patients included deflation (15), requesting larger implants (3), contracture (2), and bacterial infection (1).
Cohort study.
From 1 May through 30 September 2000, 121 women were operated on by surgeon X. Becoming a case patient was associated with OR time, time under anesthesia or surgery, having a periareolar incision, and undergoing additional surgical procedures. Women undergoing surgery in OR 2 were at a higher risk of contamination with Curvularia species (4/49 vs. 1/72; relative risk, 5.9 [95% CI, 0.751.0]; P = .16). When we conducted multivariate analyses of the data, independent risk factors associated with becoming a case patient included OR time (P = .005) and having surgery performed in OR 2 (adjusted OR, 5.2; P = .15).
Review of Practices and Procedures
Surgical procedures and practice review.
Before 1 February 2001, surgeons X and Y used the "open bowl" technique to fill the silicone shell of the inflatable saline breast implant. A sterile bowl was placed on the instrument tray before the woman arrived in the OR. Sterile saline for irrigation was poured from a bottle into the bowl, where it was exposed to the air and environment until it was drawn into a syringe and injected into the breast implant.
After 31 January 2001, when the surgeons became aware of the first 2 case patients, surgeon X started using a closed system, whereas Surgeon Y began covering the open bowl with a sterile drape. Surgeon X transferred his practice to medical center B in April 2001, and surgeon Y retired on 28 June 2001 (figure 2).
Health care worker assessment and survey.
Nineteen (56%) of 34 staff members reported having sinus problems in August 2001; all but 1 had sinus problems during MaySeptember 2000. Two staff members reported fungal infection of fingernails, and 3 reported fungal infection of toenails. Eight (24%) staff members wore artificial fingernails. Twenty-one (62%) were gardeners, and 24 (71%) had handled soil during leisure time.
Facility A Review
Operating rooms.
Facility A contained 4 ORs. A warming cabinet for saline bottles was located directly opposite and 6 feet away from OR 2. This cabinet was opened 6070 times per day.
Sterile supply room.
Before 1 July 2001, saline bottles were stored in the sterile supply room on a metal grille shelf, directly under a portion of sheetrock ceiling that had sustained water damage in 1998 following a roof leak. Portions of the sheetrock ceiling had moisture content of 28%, probably because of an ongoing leak from the roof.
Review of the HVAC system.
The HVAC system was replaced in late May/early June 2000. There was no moisture in the air ducts. According to the OR design specifications, return air should have had a flow rate of 800 cubic feet/min (cfm) and an exhaust air rate of 190 cfm, resulting in a net positive pressure in the OR. There was no documentation to demonstrate that the HVAC was balanced or to confirm the direction of airflow after the new HVAC was installed in June 2000. On 6 August 2001, the inflow rate of the supply air to OR 2 was 1235 cfm, and the outflow was 1520 cfma deficit of 285 cfm. Thus, OR 2 was maintained at negative pressure to the surrounding area; flutter strips demonstrated airflow from the corridor into the OR. OR 3 also was at negative pressure, whereas OR 1 and OR 4 were at positive pressure to the surrounding areas.
Microbiologic Studies
Personnel cultures.
Cultures were obtained from 34 facility A and 60 medical center B staff members. C. lunata was isolated from nasal mucus in 12 (35%) facility A versus 4 (6.7%) medical center B staff members (prevalence ratio, 5.3 [95% CI, 1.915.1]; P < .001). One of 12 facility A staff members with positive nasal cultures also had hand cultures positive for C. lunata, and another had a positive toenail clipping culture. Although C. lunata was isolated from the nasal mucus of surgeon X, his hand and settle plate cultures were negative. C. lunata was not isolated from surgeon Y. Isolation of C. lunata from facility A staff members was not statistically associated with sinus symptoms or exposure to plants or soil.
Environmental cultures.
C. lunata was isolated from 2 of 18 air samples. One was obtained in the sterile supply room under the water-damaged ceiling, where the saline bottles were stored, and 1 was obtained from the corridor between OR 2 and the warming cabinet. Each air sample filtered 500 L of air. None of the other 179 environmental cultures grew Curvularia species. The number of fungal colony-forming units per cubic meter of air obtained in OR 2 was 66, versus 8 in OR 3; all other conditions were identical.
DISCUSSION
The factors that contributed to contamination of SALBIs with Curvularia species among women who underwent CBA surgery at facility A included the open bowl method, which led to contamination of the saline, and engineering conditions that promoted fungal growth and contamination of the sterile supply room and ORs. Having surgery performed in OR 2 and OR time were associated with increased risk of contamination of implanted SALBIs with Curvularia species.
The saline used to fill the silicone shells was likely contaminated while sitting in an open bowl before or during surgery. Our investigation suggests that the silicone shells themselves were not contaminated at the manufacturer for the following reasons: (1) the SALBI shells had different lot numbers and were produced by 2 manufacturers, and (2) the SALBI of the first case patient became infected only after the saline of 5-year-old, previously macroscopically clear SALBIs was replaced. In 1995, Young et al. were concerned that the open bowl method might have resulted in contamination of injectable saline [8]. Although both the McGhan and Mentor Corporations provide a closed filling system, such a system is not actively promoted.
More than 7.4 million persons in the United States underwent cosmetic plastic surgery during 2000; 2.5% of these were women who underwent CBA surgery [1]. An additional 78,832 women underwent breast reconstructive surgery; 75% of implants used in these procedures were SALBIs [2]. This, however, might be an underestimate of the actual number of CBA surgical procedures performed annually, since increasing numbers of nonsurgeons are now performing cosmetic surgical procedures [12]. In most states today, increasing numbers of women are electing to have their CBA surgery performed under local anesthesia (e.g., intercostal nerve block) in outpatient clinics, ambulatory surgery centers, or physician offices, where one would expect the risks for complications, including infection, to be substantial. In fact, there have been a few case reports of fungal contamination involving SALBIs; these fungi include Aspergillus species, Curvularia species, and Paecilomyces species [810]. In general, however, infection after augmentation mammoplasty remains uncommon.
Our investigation showed that long OR time and duration of surgery were risk factors for contamination with Curvularia species. The surgical technique that uses the periareolar surgical incision approach takes longer than do other incisional approaches for CBA surgery. In addition, women who elected to undergo other procedures at the same time as their CBA surgery would have spent a longer time in the OR. Thus, undergoing CBA surgery involving the periareolar incision approach and undergoing additional procedures at the same time as CBA represent surrogate markers for a longer-than-anticipated OR time; hence, their significant association with SALBI contamination during the study period. The strong association with OR time, a marker for duration of exposure of saline in the open bowl to the environment, suggests that fungal spores in the OR air or environment extrinsically contaminated the saline.
We constructed a hypothesis to explain the sequential mode of SALBI contamination: (1) the moist sheetrock ceiling in the sterile supply room provided favorable growth conditions for C. lunata; (2) fungal spores, via air currents, dust, or water droplets, settled from the ceiling onto the surface of the saline bottles stored directly under the water-damaged ceiling; and (3) the contaminated saline bottles then were placed into the warming cabinet, where the constant opening and closing of the cabinet door resulted in air drafts laden with fungal spores. The airborne fungal spores then were drawn into OR 2, which was at profound negative pressure and was located directly opposite the warming cabinet. This scenario would explain why women who underwent their surgical procedure in OR 2 were the ones most at risk of having their SALBIs contaminated with C. lunata. According to the American Institute of Architects Academy of Architecture for Health and the CDC's guidelines for prevention of surgical site infections, air movement relative to surrounding areas should be outward from ORs [13, 14]. On the basis of the findings of the investigation and the plausibility of our hypothesis, we made several recommendations aimed at controlling the present outbreak and preventing recurrences (see the Appendix). These recommendations were immediately implemented by the facility A administration.
In addition, facility A contacted 384 women by registered mail; all had SALBI insertions between April 2000 and January 2001 at facility A. Women were counseled and offered revision surgery at no cost. As of January 2005, 224 women had undergone revision surgery; implants were visually inspected, and implant fluid was sent to the CDC for fungal culture. Two contaminated breast implants were identified: 1 from the other breast of case patient 2 and 1 from a woman whose surgery was performed in October 2000. Curvularia species were not isolated from implant fluid that was macroscopically clear. The attack rate for SALBI contamination was 3.75% for the risk period of June 2000January 2001.
In this outbreak, personnel nasal cultures were a useful adjunct to high-volume air samplers, since each staff member's nose effectively filtered 3000 L of air over an 8-h shiftthat is, 6-fold the volume of 1 air sampler. Personnel cultures reflect the large volume of air filtered, as well as exposure to prior environmental conditions. There are no other published data on the prevalence or persistence of nasal colonization by Curvularia species in this region of the United States.
Our investigation had some limitations. First, women who were asymptomatic and whose SALBI contamination status therefore was not known were presumed to be noncase patients for the cohort study analysis. This presumption resulted in potential misclassification of cases and reduced power. Second, the limited number of revision surgical procedures in women operated on by surgeon Y during MaySeptember 2001 resulted in the potential for major case-ascertainment bias and prevented us from determining whether surgeon X was linked epidemiologically to the transmission of Curvularia species.
In summary, a combination of factors, such as poor moisture control in the sterile supply room that facilitated Curvularia spores in the facility A environment, the maintenance of the OR at negative rather than positive air pressure differentials, and the absence of a closed-system protocol for injecting saline into SALBIs, all contributed to SALBI contamination with Curvularia species. These findings are in keeping with health careassociated outbreaks involving other fungi, such as Aspergillus species [1517]. Ambulatory or outpatient surgical centers need to (1) follow hospital recommendations for regular maintenance of HVAC systems and balancing of airflow in ORs; (2) follow infection control guidelines; and (3) include infection control staff in all stages of planning, construction, and renovation of health care facilities and HVAC systems. This outbreak investigation provides scientific evidence to support the recommendation that ORs should be at positive pressure relative to the surrounding areas and that a closed system should be used when injecting sterile saline into SALBIs.
Acknowledgments
We are grateful to the clinicians, patients, health care workers, and public health staff at facility A who assisted in this investigation. Also, we thank the facility A administration for implementing the recommendations that followed the investigation.
APPENDIX
Recommendations
1. Use only a closed system to fill saline-filled silicone breast implants; do not use open systems, such as the "open bowl" method.
2. Balance the heating, ventilation, and air-conditioning (HVAC) system in each operating room and provide infection control and administrative staff with documentation that balancing has been performed whenever adjustments to the air-handling unit are made or the operating rooms (ORs) undergo structural renovations [13].
3. Measure and document direction of airflow from the OR to the surrounding areas on a weekly basis and each time the HVAC system is balanced [13]. A flutter strip can be used to assess the direction of airflow across entrances to ORs.
4. Remove and replace all water-damaged or moist building materials. Correct roof leaks and address any other causes of water damage.
5. Involve infection control staff in all stages (including planning) of construction, demolition, and renovation of health care facility structures and the HVAC system [13].
References
1. American Society of Plastic Surgeons. 2000 breast surgery statistics. Available at: http://www.plasticsurgery.org/public_education/loader.cfmurl=/commonspot/security/getfile.cfm&PageID=1931. Accessed 17 May 2005. First citation in article
2. American Society of Plastic Surgeons. 2000 quick facts on cosmetic and reconstructive plastic surgery trends. Available at: http://www.plasticsurgery.org/public_education/loader.cfmurl=/commonspot/security/getfile.cfm&PageID=1934. Accessed 17 May 2005. First citation in article
3. Young VL, Watson ME. Breast implant research: where we have been, where we are, where we need to go. Clin Plast Surg 2001; 28:45183. First citation in article
4. McGhan Medical. Saline-filled breast implant surgery: making an informed decision. Santa Barbara, CA: McGhan Medical, 2000. First citation in article
5. Mentor Corporation. Saline-filled breast implant surgery: making an informed decision. Santa Barbara, CA: Mentor Corporation, 2000. First citation in article
6. Cunningham BL, Lokeh A, Gutowski KA. Saline-filled breast implant safety and efficacy: a multicenter retrospective review. Plast Reconstr Surg 2000; 105:214351. First citation in article
7. Gutowski KA, Mesna GT, Cunningham BL. Saline-filled breast implants: a Plastic Surgery Educational Foundation multicenter outcomes study. Plast Reconstr Surg 1997; 100:101927. First citation in article
8. Young VL, Hertl MC, Murray PR, Lambros VS. Paecilomyces variotii contamination in the lumen of a saline-filled breast implant. Plast Reconstr Surg 1995; 96:14304. First citation in article
9. Truppman ES, Ellenby JD, Schwartz BM. Fungi in and around implants after augmentation mammaplasty. Plast Reconstr Surg 1979; 64:8046. First citation in article
10. Williams K, Walton RL, Bunkis J. Aspergillus colonization associated with bilateral silicone mammary implants. Plast Reconstr Surg 1983; 71:2601. First citation in article
11. Petersen NJ, Collins DE, Marshall JH. A microbiological assay technique for hands. Health Lab Sci 1973; 10:1822. First citation in article
12. Sarwer DB, Nordmann JE, Herbert JD. Cosmetic breast augmentation surgery: a critical overview. J Womens Health Gend Based Med 2000; 9:84356. First citation in article
13. American Institute of Architects Academy of Architecture for Health, Facilities Guidelines Institute, US Department of Health and Human Services. Guidelines for design and construction of hospital and health care facilities. Washington, DC: American Institute of Architects, 2001:vi176. First citation in article
14. Mangram AJ, Horan TC, Pearson ML, Silver LC, Jarvis WR. Guideline for prevention of surgical site infection, 1999. Hospital Infection Control Practices Advisory Committee. Infect Control Hosp Epidemiol 1999; 20:25078. First citation in article
15. Thio CL, Smith D, Merz WG, et al. Refinements of environmental assessment during an outbreak investigation of invasive aspergillosis in a leukemia and bone marrow transplant unit. Infect Control Hosp Epidemiol 2000; 21:1823. First citation in article
16. Arnow PM, Sadigh M, Costas C, Weil D, Chudy R. Endemic and epidemic aspergillosis associated with in-hospital replication of Aspergillus organisms. J Infect Dis 1991; 164:9981002. First citation in article
17. Fox BC, Chamberlin L, Kulich P, Rae EJ, Webster LR. Heavy contamination of operating room air by Penicillium species: identification of the source and attempts at decontamination. Am J Infect Control 1990; 18:3006. First citation in article