The control of press cleaning solvent vapors in a small lithographic printing establishment.

Small businesses frequently have inadequate in-house expertise to solve a variety of safety and health problems. The National Institute for Occupational Safety and Health (NIOSH) has therefore conducted a demonstration project in the commercial lithographic printing industry, which consists largely of small companies, in an effort to establish suitable control technology for airborne solvent vapors released primarily during press cleaning operations. These solvent vapors have a number of potential adverse health effects, including narcosis, kidney and liver damage, and cancer. Also, airborne anti-offset powder is a potential allergic sensitizer and cause of occupational asthma. As a means of controlling worker exposures to the vapors and dust, a local exhaust inlet was attached to the side of the press adjacent to the paper delivery point. Tempered outside air was introduced through ceiling outlets installed to make up for the exhausted air. Measurements of press operator exposure and area concentrations of solvent vapors and area concentration of anti-offset powder were made before and after installation of the new ventilation controls. Vapor concentrations were reduced by 73 percent for the press operators. Area concentrations of the vapors were reduced by 86 percent and dust concentration by 67 percent. The ventilation system was found to be suitable for vapor and dust control, although substitution of a cleaning solution containing non-carcinogenic solvents for solutions containing carcinogens was recommended.

An evaluation of a local exhaust ventilation control system for a foundry casting-cleaning operation.

A study was conducted to evaluate the effectiveness of a local exhaust ventilation system for a foundry casting-cleaning operation in which a worker cleaned gray iron castings using a variety of handheld chipping and grinding tools. The operation originally had an exhaust system consisting only of an exhaust duct terminating approximately 1 m (3 ft) above the floor and 2 m (6 ft) from the casting-cleaning workstation. An earlier evaluation of this original control system found time-weighted average exposures to respirable silica ranging from 124 to 160 micrograms/m3. The local exhaust ventilation system evaluated in this present study consisted of a downdraft booth outfitted with a turntable for manipulating the castings. The modified local exhaust ventilation system was installed at this facility and connected to the existing plant exhaust ventilation system through the original ductwork. A direct-reading instrument was used to measure the operator's respirable aerosol exposure concentrations during a single day both before and after the installation of the new workstation. The same worker was sampled both times. The operator's activities were recorded on videotape so that the exposures associated with the various tools could be determined. While day-to-day variability could not be accounted for, depending on the type of tool used the local exhaust ventilation system reduced exposures by 59 to 79% during casting cleaning by the sampled worker when compared with the original configuration. These reductions were statistically significant.

An engineer's perspective of the intervention research workshop.

As an engineer practicing in the area of engineering controls for the reduction of occupational exposures to hazardous agents, the Intervention Research Workshop studies appeared to be effective for the studied situations, yet the models may not be applicable across a wide range of intervention possibilities. A more generic model might be more appropriate, perhaps a business model. Dr. W. Edwards Deming has proposed a model for the production of goods and services. If we can view interventions as products which need to be sold to potential customers, perhaps this type of model would be more effective across a wide range of intervention strategies.

Potential problems associated with the Rustrak Ranger data logger's data storage technique.

To maximize the 8 kilobyte memory of the Rustrak Ranger data logger, a patented technique called adaptive storage was used. With adaptive storage, the data logger predicts the value of an incoming reading. Based on previously stored readings, the data logger predicts the next reading to be constant, linear, or exponential. If the reading falls outside a window surrounding the predictions, it is stored; otherwise, it is discarded. The size of the window affects the rate at which the readings are stored, and thus is adjusted so data can be collected for the specified sampling time. If the windows are made too large, there is a resultant loss in the resolution of the data. The characteristics of the instrument and the contaminant or physical agent being measured (i.e., air contaminant versus noise) also affect data resolution. In this evaluation, previously collected exposure data (air monitoring) were sent by a computer to the Rustrak Ranger for recording. To determine the effect of the sampling time on resolution, the recorded data were analyzed using two different methods. Sampling times varied from 2 to 120 minutes. These tests showed poor resolution for sampling times as short as 30 minutes, indicating that the Rustrak Ranger may not be suitable for certain types of sampling schemes. However, if used within its limitations, this data logger is a valuable tool for recording real-time industrial hygiene data.

A hazard and operability study of anhydrous ammonia application in agriculture.

Researchers from the National Institute for Occupational Safety and Health (NIOSH) applied Hazard and Operability (HAZOP) analysis to examine hazards during the use of anhydrous ammonia by farmers. This analysis evaluated the storage, transfer, and application of anhydrous ammonia, identifying credible hazard scenarios, practical solutions, and research needs. Ninety-five findings were developed that are of use to farmers, distributors of ammonia and application equipment, and manufacturers of application equipment. The findings generally involve training, equipment design changes, preventive maintenance, and material compatibilities. The HAZOP team found that additional safety features need to be developed or implemented. The study also pointed out where correct operator procedure and preventive maintenance can prevent inadvertent releases. Other inadvertent releases are caused by incompatible materials, or by using equipment in ways other than intended. Several examples of the findings are given to emphasize the HAZOP technique and the high-risk scenarios. Strategies for dissemination to the agricultural community are presented.

Successful reduction of silica exposures at a sanitary ware pottery.

Researchers from the National Institute for Occupational Safety and Health (NIOSH) conducted a joint survey with the New Jersey Department of Health (NJDOH) to measure crystalline silica exposures and evaluate the adequacy of the existing control measures for reducing these exposures at a sanitary ware pottery. This survey found that 95% of the personal and area samples from the Slip House, Casting, Glaze Spray, and Glaze Preparation Departments exceeded the NIOSH Recommended Exposure Level (87% exceeded the Occupational Safety and Health Administration Permissible Exposure Level) for crystalline silica. Three years later, a follow-up survey found statistically significant reductions in respirable crystalline silica exposures in two of four plant departments, and statistically significant reductions in area concentrations in all four plant departments. These reductions were accomplished through a combination of automating and enclosing the batching system in the Slip House and by replacing the mold parting compound with a nonsilica material, altering the method of dry sweeping, cleaning of castings while damp, improving exhaust ventilation at the spray booths, and improved housekeeping.

Reducing exposures during the pouring operations of a brass foundry.

The focus of this exposure assessment and control technology study was a brass foundry and the lead exposures of workers involved in the transportation and pouring of metal. Controls in place at the foundry included ventilation systems at the furnace and along the continuous and stationary pouring lines. Real-time measurements were made to determine which tasks were the primary exposure sources, and a hand-held aerosol monitor was used to measure real-time aerosol exposures (as a surrogate for lead) in the workers' breathing zones. Data were collected over two 30-min sampling periods while worker activities were monitored using a video camera. Analysis of the data showed that the greatest aerosol exposures occurred during the transportation of an unventilated, full ladle, resulting in an average concentration of at least twice that of the other tasks. The study concluded that the addition of exhaust ventilation such as a moveable hood and duct system during the ladle transport and pouring tasks, and the implementation of a side draft hood at the pigging area, could result in a reduction of worker exposure to aerosols during the continuous pouring operation by up to 40%. The controls and techniques suggested in this study could be applied to pouring operations throughout the industry to reduce worker exposure to metal fumes.

Silica exposure in hand grinding steel castings.

Exposure to silica dust was studied in the grinding of castings in a steel foundry that used conventional personal sampling methods and new real-time sampling techniques developed for the identification of high-exposure tasks and tools. Approximately one-third of the personal samples exceeded the National Institute for Occupational Safety and Health recommended exposure limit for crystalline silica, a fraction similar to that identified in other studies of casting cleaning. Of five tools used to clean the castings, the tools with the largest wheels, a 6-in. grinder and a 4-in. cutoff wheel, were shown to be the major sources of dust exposure. Existing dust control consisted of the use of downdraft grinding benches. The size of the casting precluded working at a distance close enough to the grates of the downdraft benches for efficient capture of the grinding dust. In addition, measurements of air recirculated from the downdraft benches indicated that less than one-half of the respirable particles were removed from the contaminated airstream. Previous studies have shown that silica exposures in the cleaning of castings can be reduced or eliminated through the use of mold coatings, which minimize sand burn-in on the casting surface; by application of high-velocity, low-volume exhaust hoods; and by the use of a nonsilica molding aggregate such as olivine. This study concluded that all these methods would be appropriate control options.

An overview of process hazard evaluation techniques.

Since the 1985 release of methyl isocyanate in Bhopal, India, which killed thousands, the chemical industry has begun to use process hazard analysis techniques more widely to protect the public from catastrophic chemical releases. These techniques can provide a systematic method for evaluating a system design to ensure that it operates as intended, help identify process areas that may result in the release of a hazardous chemical, and help suggest modifications to improve process safety. Eight different techniques are discussed, with some simple examples of how they might be applied. These techniques include checklists, "what if" analysis, safety audits and reviews, preliminary hazard analysis (PHA), failure modes and effect analysis (FMEA), fault tree analysis (FTA), event tree analysis (ETA), and hazard and operability studies (HAZOP). The techniques vary in sophistication and scope, and no single one will always be the best. These techniques can also provide the industrial hygienist with the tools needed to protect both workers and the community from both major and small-scale chemical releases. A typical industrial hygiene evaluation of a facility would normally include air sampling. If the air sampling does detect a specific hazardous substance, the source will probably be a routine or continuous emission. However, air sampling will not be able to identify or predict the location of a nonroutine emission reliably. By incorporating these techniques with typical evaluations, however, industrial hygienists can proactively help reduce the hazards to the workers they serve.

The influence of decreased retinal illumination on automated perimetric threshold measurements.

Decreased retinal illumination (such as can be caused by pupillary constriction or light absorption by ocular media opacities) was simulated with a randomly ordered series of neutral density filters in front of the right eyes of five subjects with dilated pupils. Threshold measurements were performed on Humphrey and Octopus perimeters at 0, 5, 10, 15, 20, and 25 degrees nasally along the 180-degree meridian. A 0.6-log unit neutral density filter, which reduces retinal illumination the equivalent of halving the pupillary diameter, decreased the mean Humphrey thresholds by 1.1 +/- 0.8 decibels (dB) (mean +/- standard deviation) and the mean Octopus thresholds by 1.7 +/- 1.4 dB. Statistically significant (P less than or equal to .05, Dunnett's test) threshold depressions were observed at all eccentricities with a 1.5-log unit neutral density filter on the Humphrey perimeter (-4.5 +/- 0.7 dB) and with a 1.0-log unit neutral density filter on the Octopus perimeter (-3.5 +/- 1.0 dB).

The influence of simulated light scattering on automated perimetric threshold measurements.

The effect of light scattering by ocular media opacities on OCTOPUS and Humphrey perimeter threshold measurements was simulated with randomly ordered sequences of six ground-glass diffusers in the right eyes of five subjects. Threshold measurements were performed at 0 degrees, 5 degrees, 10 degrees, 15 degrees, 20 degrees, and 25 degrees nasally along the 180 degrees meridian with the F4 program on an OCTOPUS perimeter, and with twice-repeated profiles on a Humphrey perimeter. The reduction in differential light sensitivity correlated well with the 2.7-dB to 16.7-dB reduction in stimulus intensity caused by the 46% to 98% scattering of incident light by the diffusers. Contrast sensitivity in the presence of a glare source (Miller-Nadler glare test) was also affected by the diffusers, such that a 75% contrast target was not visible through the strongest diffuser whereas a 5% contrast target was visible without a diffuser. Conversely, the diffusers had very little effect on visual acuity measurements that were performed with projected high-contrast targets in a darkened room. Our data suggest that even minimal light scattering, such as might be caused by a cataract that has a relatively insignificant effect on visual acuity, may influence threshold measurements.

The influence of refraction accuracy on automated perimetric threshold measurements.

The effect of refraction accuracy on the differential light threshold of a Goldmann size III (0.43 degrees diameter) stimulus was evaluated in the right eyes of five normal subjects with randomly ordered sequence of plano, +1.00-, +2.00-, +3.00-, +4.00-, +5.00-, and +6.00-diopter (D) spherical overcorrections. Threshold measurements were performed at 0 degree, 5 degrees, 10 degrees, 15 degrees, 20 degrees, and 25 degrees eccentricity nasally along the 180 degrees meridian with the F4 program on the Octopus 201 perimeter. The +1.00- through +6.00-D overcorrections decreased the threshold values by an average of 1.4, 2.9, 4.2, 5.6, 6.7, and 7.6 dB, respectively, compared with the plano overcorrection. The depression of the differential light threshold was of similar magnitude at all eccentricities within the central 25 degrees.

Glaucoma filtering surgery with 5-fluorouracil.

A life-table analysis of surgical outcomes was performed on the first eye of 155 patients who were enrolled in a pilot study of glaucoma filtering surgery with postoperative subconjunctival 5-fluorouracil (5-FU) injections. The success rates at 1-, 2-, and 3-year intervals were 68, 63, and 63%, respectively, for 88 patients with non-neovascular glaucoma in aphakia; 82, 75, and 75% for 39 patients with non-neovascular glaucoma after unsuccessful filtering surgery; and 68% at each yearly interval for 28 patients with neovascular glaucoma. Complications which resulted from filtering surgery and the 5-FU injections included corneal epithelial defects (55.5%), conjunctival wound leaks (36.8%), suprachoroidal hemorrhage (5.8%), rhegmatogenous retinal detachment (2.6%), endophthalmitis and phthisis (1.9% each), and corneal scarring, late bleb leak, malignant glaucoma, and traction retinal detachment (1.3% each). A Cox Model regression analysis failed to demonstrate a correlation between surgical success and age, race, type of filtering procedure, or total dose of 5-FU received. Postoperative subconjunctival 5-FU may increase the operative success rate for selected patients with a high risk for failure after glaucoma filtering surgery.

Wound healing after filtering surgery in owl monkeys.

We performed posterior lip sclerectomies in 29 owl monkey eyes to determine the natural history of wound healing after filtering surgery without adjunctive antimetabolite therapy. We noted three phases of wound healing during clinical and histologic examination: early healing (days 2 to 6), intermediate healing (days 7 to 9), and late healing (days 10 to 14). In the early healing phase, all limbal fistulas except one remained open gonioscopically, but by day 6, fibroblasts had proliferated along the walls of the opening. Proliferation and migration of fibroblasts continued during the intermediate healing phase to completely occlude four and to partially occlude two of the ten fistulas in the eyes studied during this time. In the late healing phase, the limbal fistula was completely closed by granulation tissue in four of five eyes and was slitlike open in one eye. In this model of filtering surgery, wound healing at the sclerectomy site with obliteration of the limbal opening by proliferating fibroblasts occurred within the first 14 postoperative days. We believe that the short-term effects of newer treatments designed to alter wound healing after filtering surgery may be assessed in this model, which is characterized by predictable and prompt wound healing.

5-Fluorouracil and glaucoma filtering surgery. III. Intermediate follow-up of a pilot study.

Ninety-five patients (104 eyes) were enrolled in a pilot study of subconjunctival 5-fluorouracil (5-FU) injections after filtering surgery in eyes with poor surgical prognoses. At least a six-month follow-up was available on 84 patients, of whom four were excluded from the analysis of surgical outcome because they suffered retinal detachments within six months of their filtering surgery. The initial 5-FU procedures on the remaining 80 patients were successful (no further glaucoma surgical procedures were either performed or recommended, and the intraocular pressures [IOPs] were either 21 mmHg or lower with ocular hypotensive medication[s] or 25 mmHg or lower without ocular hypotensive medication) in 33 (68%) of the 48 aphakic eyes with non-neovascular glaucomas, 13 (81%) of the 16 phakic eyes with non-neovascular glaucomas after unsuccessful filtering surgery, and 12 (75%) of the 16 eyes with neovascular glaucoma (NVG). The follow-up on the successful eyes ranged from 6 to 34 months (mean +/- SD = 18.5 +/- 7.5). The visual acuities remained within one line of their preoperative levels or improved in 38 (79%) of the 48 aphakic eyes with non-neovascular glaucoma, 11 (69%) of the 16 phakic eyes with non-neovascular glaucoma after unsuccessful filtering surgery, and 12 (75%) of the 16 eyes with NVG. The initial 5-FU procedures on the 104 eyes were complicated by the following: corneal epithelial defects (50%); conjunctival wound and suture tract leaks (36%; 2% underwent surgical repair); suprachoroidal hemorrhages (9%); retinal detachments (3%); subepithelial corneal scarring (3%); endophthalmitis (2%); and malignant glaucoma (1%). It is the authors' impression that postoperative subconjunctival 5-FU increases the likelihood of achieving IOP control after filtering surgery in eyes with poor surgical prognoses; however, a randomized clinical trial is necessary to confirm this impression.

Topical fluorouracil. II. Postoperative administration in an animal model of glaucoma filtering surgery.

Unilateral posterior lip sclerectomies were performed in ten owl monkeys. Five milligrams of fluorouracil was injected subconjunctivally in each operated eye immediately after surgery. Three drops (approximately 2.4 mg/drop) of fluorouracil were instilled ten minutes apart in each operated eye twice daily on postoperative days 1 through 7 and once daily on postoperative days 8 through 15, 17, 19, and 21. One monkey died on the seventh postoperative day; its death could not be attributed to systemic fluorouracil toxicity. All of the operated eyes had filtering blebs after the full course of fluorouracil, but seven also had corneal epithelial defects. By the seventh postoperative week, two of the operated eyes manifested moderately severe corneal opacification. Ten weeks postoperatively, the electroretinographic a- and b-wave amplitudes averaged 17% and 12% less, respectively, in the seven operated eyes without clinically significant corneal opacification than in the unoperated fellow eyes. Only two eyes had blebs after the 12th postoperative week. Histopathologic examination was performed on five eyes, of which only two revealed patent sclerostomies. Although topical fluorouracil appears to delay bleb scarring, the corneal findings suggest that it may be more toxic than subconjunctival fluorouracil.

Topical fluorouracil. Pharmacokinetics in normal rabbit eyes.

Postoperative subconjunctival fluorouracil injections may be a useful adjunct to standard glaucoma filtering surgery in eyes that are at high risk of failure. Topical administration would be preferable to subconjunctival administration; however, there are no data on the ocular penetration of topically applied fluorouracil. Consequently, we investigated the pharmacokinetics of topically administered fluorouracil labeled with carbon 14 in normal rabbit eyes. One drop (approximately 2.4 mg) of fluorouracil resulted in the following concentrations at 0.5 and six hours, respectively: 17.3 and 0.9 micrograms/g of conjunctiva; 24.3 and 1.3 micrograms/g of cornea; 14.6 and 0.2 micrograms/mL of aqueous; 0.8 and 0.5 microgram/g of lens; 1.1 and 0.3 microgram/g of vitreous; and 0.2 and less than 0.1 microgram/mL of serum. Three drops (approximately 7.2 mg) of fluorouracil resulted in the following concentrations at 0.5 and eight hours, respectively: 589.8 and 1.3 micrograms/g of conjunctiva; 502.9 and 1.8 micrograms/g of cornea; 199.6 and 0.8 micrograms/mL of aqueous; 6.2 and 0.5 micrograms/g of lens; 6.8 and 0.5 micrograms/g of vitreous; and 1.3 and 0.2 microgram/mL of serum. Since a fluorouracil concentration of 0.2 microgram/mL inhibits rabbit conjunctival fibroblast proliferation in cell culture by 50%, these data suggest that topically applied fluorouracil achieves sufficient levels in the ocular compartments and tissues to have potential therapeutic applications.

5-Fluorouracil toxicity to the ocular surface epithelium.

The antimetabolite, 5-fluorouracil (5-FU), has been used to control proliferation of retinal pigment epithelial cells and fibrocytes, and is currently the subject of a multicenter clinical trial of its value in the control of scarring after glaucoma operations. To evaluate possible ocular surface toxicity, the effect of 5-FU on the mitotic rate and differentiation of the ocular surface epithelium in rabbits was measured. 5-FU was instilled into eyes with 10-mm diameter central epithelial wounds and into nonwounded eyes at a dose of 9 mg per day for 4 days. Saline treated control wounded eyes healed within 4 days (n = 5) while 40% (4 of 10) of the 5-FU treated wounded eyes had defects at 4 days. The normal mitotic rate of the corneal epithelium was 1.0 +/- 0.3 (n = 4) tritiated thymidine labeled cells per 100 basal corneal epithelial cells after 2.5 hr incubation. Saline treated control wounded eyes had an increased mitotic rate, 7.1 +/- 1.3 (n = 5) labeled cells per 100 basal corneal epithelial cells after 2.5 hr incubation. Topical 5-FU decreased both of those rates to about 1% of normal. The normal conjunctival epithelial mitotic rate was 1.8 +/- 0.4 (n = 4) labeled cells per 60 basal cells after per 2.5 hr incubation. This rate was the same in wounded eyes, but was decreased in eyes treated with 5-FU. Thus, 5-FU (9 mg/day topically) has serious toxic effects to ocular surface epithelium which must be carefully considered if this drug is to be used clinically.

Delayed nonexpulsive suprachoroidal hemorrhage.

Eight cases of delayed nonexpulsive suprachoroidal hemorrhage occurred after filtering operations in aphakic eyes. Sustained hypotonia resulting in serous choroidal detachments, scleral infolding, or both was observed prior to the hemorrhage in seven of the eight eyes. The prognosis was related to the extent of the hemorrhage and possibly to whether or not surgical drainage was performed. Only one of the five patients with massive hemorrhages recovered visual acuity better than hand motions; that patient had undergone prompt surgical drainage of the suprachoroidal blood.