Cincinnati, OH: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, HETA 2001-0303-2893, 2002 Oct; :1-36
On May 11, 2001, the National Institute for Occupational Safety and Health (NIOSH) received a request from the United Automobile, Aerospace, and Agricultural Implement Workers of America (UAW) Local 1939 and management at the TRW Automotive plant in Mt. Vernon, Ohio, to conduct a health hazard evaluation (HHE). The request concerned respiratory problems, including hypersensitivity pneumonitis (HP), thought to be associated with occupational exposures to metalworking fluids (MWFs). Prior to receipt of the request, on May 9, 2001, a meeting had been held at NIOSH Hamilton Laboratories in Cincinnati, Ohio, with representatives of NIOSH, TRW, and UAW Local 1939 and UAW Health and Safety. At that meeting, respiratory symptoms and illnesses among workers were discussed. It was reported that symptoms began in approximately October 2000, and that five workers had subsequently been diagnosed by their personal physicians with HP. Subsequent evaluations and actions taken by TRW responses were also discussed. Among the industrial hygiene and medical issues discussed at the meeting, it was noted that atypical mycobacteria (Mycobacterium chelonae) had been cultured from the MWF (along with other bacteria) and that several of the workers first diagnosed with HP had been found to have antibodies to the M. chelonae on precipitin testing. Subsequent to the meeting and the HHE request, multiple site visits were made to the Mt. Vernon plant. Environmental evaluations conducted by NIOSH investigators occurred in June and December 2001. The industrial hygiene evaluations included records reviews, personal breathing zone (PBZ) and area air sampling during usual operations as well as during cleaning operations, and bulk sampling of MWF. Medical evaluation included medical record review, discussions with employees' private physicians, and a questionnaire survey. Initial microbial culture data collected by TRW, and confirmed by subsequent NIOSH sampling, revealed that the central MWF systems were contaminated with up to 10(7) colony forming units per milliliter (CFU/ml) total bacteria and 105 CFU/ml Mycobacterium species. In May 2001, the acid fast bacteria (AFB) pellet stain (used to assess the quantity of dead and living Mycobacteria) revealed 'very high' and 'high' concentrations in two of the four systems. By mid-August, about six weeks after TRW began using a new biocide (para-chloro, meta-cresol [PCMC]), no fungal or bacterial growth was found from process MWF samples. Repeated AFB pellet stains revealed a slow downward trend in the concentration of Mycobacteria levels over the course of the HHE. Air sampling data (a mixture of area and PBZ samples) collected by TRW from June 2001 to February 2002 were reviewed. The mean concentrations of total particulate for these groups of samples by department ranged from 0.14 to 0.68 mg/m3. Sixteen PBZ MWF aerosol samples for thoracic fraction of MWF aerosol taken by NIOSH representatives during normal production on June 29, 2001, ranged from below the limit of detection to 0.37 mg/m3 (mean concentration was 0.23 mg/m3). All samples contained a large percentage of extractable material, indicating that the exposures were primarily MWF aerosol. Aerosol sampling conducted during cleanup operations beginning the night of June 29, 2001, revealed airborne thoracic particulate concentrations inside the containment area ranging from 0.13 mg/m3 to 0.51 mg/m3 (mean: 0.41 mg/m3). Concentrations outside the containment area ranged from 0.03 mg/m3 to 0.2 mg/m3 (mean: 0.1 mg/m3). PBZ sampling for MWF aerosol on December 18, 2001, after air-conditioning had been added to the machining areas, revealed concentrations ranging from 0.14 mg/m3 to 0.69 mg/m3. Five exposures were above the NIOSH REL for thoracic particulate (0.4 mg/m3); all five were from operators who worked on a particular horizontal broach machine as a part of their job rotation. On May 21-22, 2001, a questionnaire was administered to all employees at the plant. Three hundred five (91%) of 335 employees completed the questionnaire, which included questions concerning respiratory and systemic symptoms occurring within the year prior to the survey. All symptoms included in the questionnaire were reported more frequently among those employees exposed to MWF compared to those who were not exposed, with prevalence ratios for symptoms ranging from 1.2 - 2.2. All but one of the 95% confidence intervals (CI) for the prevalence ratios excluded 1, suggesting that these differences were not due to chance. Twenty (77 %) of the 26 persons meeting the case definitions for HP or occupational asthma (OA) in the November 2001 medical record review (see below) participated in the questionnaire survey. Both HP and OA were more common among the exposed group compared to the unexposed group, although the CI for both illnesses included 1. Records reviewed in November 2001 revealed that 107 workers had been placed on work restriction due to respiratory conditions in the previous 11 months. Among these 107, 70 (65%) had returned to work as of November 2001; 37 (35%) of the total 107 remained on medical leave of absence. Medical records were reviewed for 32 (86%) of the 37 workers. Among these 32 workers, 25 (78%) had primary work duties in the machining areas of the plant, although all 32 reported intermittent work duties in the machining side of the plant. Date of onset of symptoms for these workers ranged from October 2000 to April 2001, with the majority of persons reporting onset of respiratory symptoms in December 2000 and January 2001. Among the 32 workers, 14 (44%) met a case definition for OA and 12 (38%) a definition for HP. TRW records from March 2002 revealed that 35 workers were on medical leave of absence; most of these workers had remained on medical leave of absence since November 2001. Continuing review of records and discussions with the primary treating physicians in April 2002 revealed clinical diagnoses of HP and/or OA among 41 workers, with onset of symptoms prior to May 2001. Concerns related to tremor occurring among some workers were raised during the course of the HHE. Review of medical records for seven employees seen by a private neurologist and review of the toxicity of the materials and substances used by these workers did not reveal a neurotoxic agent to which workers may be exposed at TRW. Specifically concerning PCMC (which has not been identified in the medical literature as an agent causing tremor in the workplace), four of the seven workers for whom the date of onset of tremor was reported had onset of tremor prior to the first use of PCMC at TRW. Many TRW Mt. Vernon employees have experienced a spectrum of work-related respiratory illness, with onset of symptoms occurring between the fall of 2000 and April 2001. No employees with onset of respiratory symptoms after that time have been identified. The majority of air samples revealed concentrations of MWF aerosol in the machining areas of the plant to be below the NIOSH REL, but NIOSH air sampling revealed several instances where PBZ exposures exceeded the NIOSH REL during normal operations. Air sampling during cleanup operations indicated exposures to MWF aerosol at concentrations above the NIOSH REL. Although the exact cause of employees' symptoms and illnesses has not been determined, evidence from similar outbreaks of illnesses at other workplaces suggests that contamination of the MWF with Mycobacteria sp. is playing a role. Multiple interventions were put in place over time at TRW and we are not able to identify one specific control measure which has been primarily effective in eliminating the source of the illnesses. Recommendations are provided in this report to assist TRW, the union, and workers at TRW in addressing occupational exposures in the machining areas and work-related illnesses and symptoms.