Assessment of inspiration and technical quality in anteroposterior thoracic radiographs using machine learning

INTRODUCTION

Chest radiographs are the most performed radiographic procedure, but suboptimal technical factors can impact clinical interpretation. A deep learning model was developed to assess technical and inspiratory adequacy of anteroposterior chest radiographs.

METHODS

Adult anteroposterior chest radiographs (n = 2375) were assessed for technical adequacy, and if otherwise technically adequate, for adequacy of inspiration. Images were labelled by an experienced radiologist with one of three ground truth labels: inadequate technique (n = 605, 25.5 %), adequate inspiration (n = 900, 37.9 %), and inadequate inspiration (n = 870, 36.6 %). A convolutional neural network was then iteratively trained to predict these labels and evaluated using recall, precision, F1 and micro-F1, and Gradient-weighted Class Activation Mapping analysis on a hold-out test set. Impact of kyphosis on model accuracy was assessed.

RESULTS

The model performed best for radiographs with adequate technique, and worst for images with inadequate technique. Recall was highest (89 %) for radiographs with both adequate technique and inspiration, with recall of 81 % for images with adequate technique and inadequate inspiration, and 60 % for images with inadequate technique, although precision was highest (85 %) for this category. Per-class F1 was 80 %, 81 % and 70 % for adequate inspiration, inadequate inspiration, and inadequate technique respectively. Weighted F1 and Micro F1 scores were 78 %. Presence or absence of kyphosis had no significant impact on model accuracy in images with adequate technique.

CONCLUSION

This study explores the promising performance of a machine learning algorithm for assessment of inspiratory adequacy and overall technical adequacy for anteroposterior chest radiograph acquisition.

IMPLICATIONS FOR PRACTICE

With further refinement, machine learning can contribute to education and quality improvement in radiology departments.

Groundwater Ecosystems Vary with Land Use across a Mixed Agricultural Landscape

Changes in surface land use may threaten groundwater quality and ecosystem integrity, particularly in shallow aquifers where links between groundwater and surface activities are most intimate. In this study we examine the response of groundwater ecosystem to agricultural land uses in the shallow alluvial aquifer of the Gwydir River valley, New South Wales, Australia. We compared groundwater quality and microbial and stygofauna assemblages among sites under irrigated cropping, non-irrigated cropping and grazing land uses. Stygofauna abundance and richness was greatest at irrigated sites, with the composition of the assemblage suggestive of disturbance. Microbial assemblages and water quality also varied with land use. Our study demonstrates significant differences in the composition of groundwater ecosystems in areas with different surface land use, and highlights the utility of groundwater biota for biomonitoring, particularly in agricultural landscapes.

A comparison of mixture toxicity assessment: examining the chronic toxicity of atrazine, permethrin and chlorothalonil in mixtures to Ceriodaphnia cf. dubia

Pesticides predominantly occur in aquatic ecosystems as mixtures of varying complexity, yet relatively few studies have examined the toxicity of pesticide mixtures. Atrazine, chlorothalonil and permethrin are widely used pesticides that have different modes of action. This study examined the chronic toxicities (7-d reproductive impairment) of these pesticides in binary and ternary mixtures to the freshwater cladoceran Ceriodaphnia cf. dubia. The toxicity of the mixtures was compared to that predicted by the independent action (IA) model for mixtures, as this is the most appropriate model for chemicals with different modes of action. Following this they were compared to the toxicity predicted by the concentration addition (CA) model for mixtures. According to the IA model, the toxicity of the chlorothalonil plus atrazine mixture conformed to antagonism, while that of chlorothalonil and permethrin conformed to synergism. The toxicity of the atrazine and permethrin mixture as well as the ternary mixture conformed to IA implying there was either no interaction between the components of these mixtures and/or in the case of the ternary mixture the interactions cancelled each other out to result in IA. The synergistic and antagonistic mixtures deviated from IA by factors greater than 3 and less than 2.5, respectively. When the toxicity of the mixtures was compared to the predictions of the CA model, the binary mixture of chlorothalonil plus atrazine, permethrin plus atrazine and the ternary mixture all conformed to antagonism, while the binary mixture of chlorothalonil plus permethrin conformed to CA. Using the CA model provided estimates of mixture toxicity that did not markedly underestimate the measured toxicity, unlike the IA model, and therefore the CA model is the most suitable to use in ecological risk assessments of these pesticides.

3D time-resolved MR angiography (MRA) of the carotid arteries with time-resolved imaging with stochastic trajectories: comparison with 3D contrast-enhanced Bolus-Chase MRA and 3D time-of-flight MRA

BACKGROUND AND PURPOSE

Time-resolved MR angiography (MRA) offers the combined advantage of large anatomic coverage and hemodynamic flow information. We applied parallel imaging and time-resolved imaging with stochastic trajectories (TWIST), which uses a spiral trajectory to undersample k-space, to perform time-resolved MRA of the extracranial internal carotid arteries and compare it to time-of-flight (TOF) and high-resolution contrast-enhanced (HR) MRA.

MATERIALS AND METHODS

A retrospective review of 31 patients who underwent carotid MRA at 1.5T using TOF, time-resolved and HR MRA was performed. Images were evaluated for the presence and degree of ICA stenosis, reader confidence, and number of pure arterial frames attained with the TWIST technique.

RESULTS

With a consensus interpretation of all sequences as the reference standard, accuracy for identifying stenosis was 90.3% for TWIST MRA, compared with 96.0% and 88.7% for HR MRA and TOF MRA, respectively. HR MRA was significantly more accurate than the other techniques (P < .05). TWIST MRA yielded datasets with high in-plane spatial resolution and distinct arterial and venous phases. It provided dynamic information not otherwise available. Mean diagnostic confidence was satisfactory or greater for TWIST in all patients.

CONCLUSION

The TWIST technique consistently obtained pure arterial phase images while providing dynamic information. It is rapid, uses a low dose of contrast, and may be useful in specific circumstances, such as in the acute stroke setting. However, it does not yet have spatial resolution comparable with standard contrast-enhanced MRA.

Comparison of the fate and toxicity of chlorpyrifos--laboratory versus a coastal mesocosm system

The widespread use of chlorpyrifos for pest control in urban and rural environments poses a risk of contamination to aquatic environments via runoff, spray drift or spillage. The aim of this study was to assess the fate of chlorpyrifos and its toxicity to common freshwater invertebrates in the laboratory and in stream mesocosms. Chlorpyrifos was rapidly lost from the test systems but the rates of loss varied considerably, such that losses in the mesocosms could not be reliably predicted from the static laboratory studies. This was likely due to the mass transport of chlorpyrifos from the mesocosm via stream flow. Chlorpyrifos was acutely toxic to all invertebrates tested with the cladoceran species (laboratory 48h LC(50) values 0.07-0.10 microg L(-1)) being most sensitive. Despite the differences in the dynamics of chlorpyrifos in the laboratory and mesocosm systems, the sensitivities of the mayfly Atalophlebia australis and the cladoceran Simocephalus vetulus were similar in the 2 systems.

Iliocaval arteriovenous fistula presenting with multiple organ failure

Iliac arteriovenous fistulas are an uncommon condition, which may be spontaneous or traumatic in nature. Such fistulas classically present with a triad of high-output cardiac failure, pulsatile abdominal mass with a bruit and unilateral leg ischaemia or venous congestion. We describe a case of an iliocaval fistula secondary to rupture of a common iliac artery aneurysm, with an unusual presentation of multiple organ failure, masquerading as sepsis. We describe the CT findings of iliocaval fistula, which was the means of diagnosis in this study.

Toxicity and bioavailability of atrazine and molinate to the freshwater fish (Melanotenia fluviatilis) under laboratory and simulated field conditions

Acute (96 h) semi-static toxicity tests were conducted by exposing the freshwater fish, Melanotenia fluviatilis, to atrazine and molinate in laboratory and river water both with and without sediment. The 96-h EC50 (imbalance) values of atrazine to M. fluviatilis ranged from 5.6 to 10.4 mg L(-1) while the corresponding values for molinate ranged from 7.9 to 14.8 mg L(-1), respectively. Atrazine was classed as having moderate toxicity while molinate had low to moderate toxicity to M. fluviatilis. Neither the presence of river water nor sediment significantly (P<0.05) reduced the bioavailability of either herbicide to M. fluviatilis. A series of other studies by the authors have found that sediment significantly (P<0.05) reduced the bioavailability of these two chemicals to a variety of organisms. Reasons for sediment having no effect for this species were examined. This lack of effect by sediment is most likely due to the relative rates of absorption into the fish and adsorption onto the sediment. However, contributions to this outcome by resuspended sediment, contaminated food and a combined effect of the herbicides and sediment could not be excluded.

Endovascular treatment of arterial mesenteric ischaemia: a retrospective review

Mesenteric ischaemia is a condition that has traditionally been managed surgically. It poses a challenging diagnostic and therapeutic problem, particularly in the acute setting. We review a small series of eight patients managed with endovascular techniques for either acute or chronic mesenteric ischaemia at The Royal Melbourne Hospital, from 1997 to 2002. We describe our results and relate these to the recent published literature regarding endovascular and surgical management of mesenteric ischaemia. Our experience confirms the valuable contribution of angioplasty and stenting in chronic mesenteric ischaemia, which compares favourably with surgery with regards to complication rates and mortality. We suggest the need for further studies to compare the long-term efficacy of endovascular techniques compared with surgery in the management of chronic mesenteric ischaemia. Furthermore, we demonstrate a role for endovascular management in acute mesenteric ischaemia, in the appropriate clinical setting.

Effect of river water, sediment and time on the toxicity and bioavailability of molinate to the marine bacterium Vibrio fischeri (Microtox)

The toxicity and bioavailability of molinate to Vibrio fischeri (Microtox((R))) were determined in both laboratory and river water in the absence and presence of sediment after 0, 24, 48, 72 and 96-h exposure. The bioavailability of molinate, expressed as 5min EC50s (bioluminescence) and their fiducial limits calculated using initial measured concentrations, to V. fischeri in laboratory water in the absence and presence of sediment ranged from 1.8 (1.7-2.1) to 3.6 (3.5-3.7) mgL(-1) and 1.3 (1.2-1.4) to 4.2 (3.5-4.5) mgL(-1), respectively. The corresponding values in river water and river water plus sediment were 1.7 (1.6-1.8) to 3.8 (3.6-4.1) and 1.3 (1.3-1.4) to 4.6 (4.2-4.9) mgL(-1), respectively. River water did not significantly (P>0.05) reduce the bioavailability of molinate to V. fischeri compared to that of laboratory water. However, the presence of sediment significantly (P<0.05) reduced the bioavailability of molinate to V. fischeri in both waters. The exposure time also significantly (P<0.05) reduced the bioavailability of molinate to V. fischeri in both waters in the presence and absence of sediment. The type of water did not significantly (P>0.05) affect the loss of molinate during the 96-h exposure period. However, the presence of sediment significantly (P<0.01) increased the loss of molinate from the test solutions, probably by binding to the sediment particles. Exposure period and concentration levels significantly (P<0.05) affected the loss of the herbicides over the 96h.

The development of marine Toxicity Identification Evaluation (TIE) procedures using the unicellular alga Nitzschia closterium

Unicellular algae are highly sensitive to a wide range of toxicants and have been used extensively in ecotoxicological testing. This, along with their ability to grow in very small test volumes over short test durations, make them ideal test organisms for use in Toxicity Identification Evaluations (TIEs). Despite this, microalgae have not previously been used in marine TIE studies. In this study, the marine diatom Nitzschia closterium was shown to be a highly suitable test organism after modification of the standard test protocol to reduce test volumes to 6 mL and test duration to 48 h. The alga was tolerant to the chemicals used in phase I of the standard USEPA TIE protocol, and physical TIE manipulations had no effect on algal growth. The cation exchange procedure, however, inhibited algal growth, while the anion exchange procedure stimulated growth, making these two procedures unsuitable for use with this species. Of the buffers trialed for the graduated pH procedure, 0.01 M PIPES buffer was found to be suitable for buffering at pH 7 because it maintained the required pH over the duration of the test and did not affect the growth or sensitivity of the algae to one reference toxicant (copper). A trial TIE on a secondary-treated sewage effluent for discharge into coastal waters showed that the developed protocols could successfully be used to identify ammonia as the major toxicant in the effluent.

The toxicity and bioavailability of atrazine and molinate to Chironomus tepperi larvae in laboratory and river water in the presence and absence of sediment

Acute (10 day) semi-static toxicity tests in which the midge, Chironomus tepperi, were exposed to atrazine and molinate were conducted in laboratory water and in river water, in the absence and presence of sediment. The bioavailability measured as median lethal concentrations (LC50) and 95% fiducial limits (FLs) of atrazine to C. tepperi in laboratory water in the absence and presence of sediment were 16.6 (14.3-19.4) and 21.0 (18.2-24.1) mg l(-1), respectively while the corresponding values in river water were 16.7 (14.7-19.0) and 22.7 (20.3-25.4) mg l(-1), respectively. For molinate, the LC50 and FL values in laboratory water in the absence and presence of sediment were 8.8 (6.8-11.4) and 14.3 (12.4-16.4) mg l(-1), respectively and the corresponding values in river water were 9.3 (7.6-11.3) and 14.5 (12.4-16.9) mg l(-1), respectively. Atrazine has low toxicity (LC50 > 10 mg l(-1)) while molinate has moderate toxicity (1 mg l(-1) < LC50 < 10 mg l(-1)) to C. tepperi. River water did not significantly (P > 0.05) reduce the bioavailability of either chemical to C. tepperi. However, the presence of sediment did significantly (P < 0.05) reduce the bioavailability of both atrazine and molinate to C. tepperi.

Toxicity and bioavailability of atrazine and molinate to the freshwater shrimp (Paratya australiensis) under laboratory and simulated field conditions

Acute (96-h) semistatic toxicity tests were conducted by exposing the freshwater shrimp, Paratya australiensis, to atrazine and molinate in laboratory water and in river water both with and without sediment. The median lethal concentrations (LC50) and 95% fiducial limits of atrazine for P. australiensis in laboratory water in the absence and presence of sediment were 9.9 (8.6-11.5) and 6.8 (5.4-8.5)mg/L, respectively, while the corresponding values in river water were 9.8 (8.5-11.2) and 6.5 (5.4-7.8)mg/L, respectively. For molinate, the LC50 values in laboratory water in the absence and presence of sediment were 9.2 (7.0-12.1) and 9.0 (6.8-12.0)mg/L, respectively and the corresponding values in river water were 8.7 (6.4-11.8) and 8.2 (6.6-10.2)mg/L, respectively. Neither the river water nor the presence of sediment significantly (P<0.05) reduced the bioavailability of either chemical to P. australiensis. This was unexpected, as studies with other aquatic organisms have shown that sediment significantly reduced the bioavailability of these chemicals.

Angioplasty and stenting of the superior mesenteric artery in acute mesenteric ischaemia

Acute mesenteric ischaemia is a challenging diagnostic problem with a very high mortality. Traditionally, laparotomy is required for definitive management. We describe a successful case of angioplasty and stenting of the superior mesenteric artery in a surgically unfit patient. We recommend that stenting be considered only in situations where the diagnosis has been made prior to bowel infarction, and if the patient poses a poor operative risk.

Toxicity of atrazine and molinate to the cladoceran Daphnia carinata and the effect of river water and bottom sediment on their bioavailability

Atrazine and molinate are widely used herbicides and concern has been raised about their potential deleterious impacts on aquatic ecosystems. Although there have been some studies on the toxicity of herbicides to aquatic organisms using laboratory or natural water, information on the effect of sediments, suspended particulates, and dissolved organic matter on their bioavailability is quite limited. This study aims to provide toxicity data that considers these factors and the effect that these factors have on bioavailability. In this study, the toxicity of the test chemicals was calculated following the Organisation for Economic Co-operation and Development (OECD) methods, whereas change in bioavailability was measured using EC50 values based on measured initial concentrations of the test chemicals. The acute (48-h) static toxicity of atrazine and molinate to the freshwater cladoceran Daphnia carinata was determined in cladoceran water and river water in the absence and presence of sediment. The 48-h EC50 (immobilization) values of atrazine to D. carinata ranged from 22.4 to 26.7 mg/L, while the corresponding values for molinate ranged from 18.3 to 33.6 mg/L, respectively. Both chemicals were classed as having low acute toxicity to D. carinata. The presence of dissolved organic matter and suspended particles in river water did not significantly (p > 0.05) reduce the bioavailability (measured as toxicity) of atrazine to D. carinata compared to that tested in cladoceran water. The presence of sediment, however, significantly (p < 0.05) reduced the bioavailability (48-h EC50) of atrazine in cladoceran water, from 24.6 to 30.7 mg/L, and in river water, from 22.4 to 31.0 mg/L. Similarly, the presence of sediment in cladoceran water, significantly (p < 0.05) reduced the bioavailability (48-h EC50) of molinate, from 26.6 to 46.4 mg/L, and in river water, from 22.5 to 45.6 mg/L.

The effects of pH and dissolved organic carbon on the toxicity of cadmium and copper to a freshwater bivalve: further support for the extended free ion activity model

The extended free ion activity model (FIAM) was developed by integrating concepts from the original FIAM into biological receptor theory, to obtain a conceptual model that more precisely quantifies the interaction of chemical species at biological receptor sites. The extended FIAM was tested by determining the acute (48 h) valve movement behavior (VMB) (measured in terms of the duration of valve opening) of the Australian freshwater bivalve, Hyridella depressa, to increasing concentrations of total Cd or Cu, in a standard synthetic water under conditions of varying pH (6.5-7.5) and/or dissolved organic carbon (as model fulvic acid (FA)) concentrations (0-11.2 mg L(-1)). Valve movement behavior, measured using an automated data acquisition system, was shown to be a quantifiable and rapid, real-time endpoint for assessing the toxic effects of Cd and Cu exposures. The VMB of H. depressa to Cd was independent (p > 0.05) of pH and/or model FA concentration. In contrast, the VMB of H. depressa to Cu was highly dependent (p < 0.001) on pH and/or model FA concentration; individuals were more sensitive to Cu at low pH and model FA concentrations. The VMB of H. depressa was directly proportional to the activity of the free metal ion (Cd2+), for the linear region of the concentration-response curves. In contrast, the VMB of H. depressa was a weighted function of the activities of the free metal ion and the 1:1 metal hydroxide species (i.e. 2.02 x Cu2+ + CuOH+), whereby Cu2+ had a two-fold greater binding affinity than CuOH+ at the cell membrane surface. Moreover, the results for Cd and Cu are consistent with the extended FIAM, as opposed to the original FIAM, where the result for Cu would be regarded as an exception. The extended FIAM explained 98% of the variability in VMB, whereas the original FIAM explained only 63% (i.e. an improvement of 35%). The improved predictability of organism response to Cu is relevant to advancing water quality guidelines for protecting aquatic biota.

Experimental study of blood lead kinetics in estuarine crocodiles (Crocodylus porosus) exposed to ingested lead shot

A previous study of lead (Pb) contamination in estuarine crocodiles (Crocodylus porosus) in Kakadu National Park, Australia, found elevated Pb levels in bone and flesh from individuals caught in habitats where hunting with lead ammunition had occurred. Lead shot was suspected as the potential source of Pb in these animals. An investigation was designed to determine if crocodiles are capable of retaining and dissolving lead shot in their stomachs and absorbing the ionic Pb into the blood. Four of the six juvenile crocodiles (C. porosus) used in the experiment were fed an initial dose of five or ten lead shot each. Most of the lead shot were retained in the stomach over the 20-week experimental period, as confirmed by stomach lavage and radiography of the stomach region. From 13 to 30% of the original weight of individual lead shot had been eroded in that time. In vitro experiments confirmed that lead shot could be dissolved under conditions simulating those known for the crocodilian stomach. Blood Pb concentrations increased by an order of magnitude within a week after dosing three crocodiles with five lead shot each, and then attained steady-state equilibrium 5-20 weeks after dosing, with blood Pb steady-state levels ranging from 278 to 363 microg/dL. The blood Pb concentration-time curves in these crocodiles followed a one-compartment model with first-order loss kinetics that yielded an apparent biological half-life for Pb in blood of about 3.4 days. Throughout the experiment, the crocodiles remained in apparent good physical condition and displayed no clinical signs of Pb toxicosis. These observations, together with the very high blood Pb levels that were sustained for several months, suggest that crocodilians may possess a relatively high degree of resistance to Pb toxicity.

Assessment of metal toxicity in sediment pore water from Lake Macquarie, Australia

Recent investigations into the level of heavy metal enrichment in the sediments of Lake Macquarie have indicated that significant contamination has occurred over the past 100 years, with elevated levels of lead, zinc, cadmium, copper, and selenium being observed in most parts of the lake. Pore water extracted from sediments showing the greatest contamination by these metals exhibited toxicity to the larval development of the sea urchin Heliocidaris tuberculata. However, an analysis of pore water metal concentrations revealed that the concentrations of these metals were too low to cause toxicity. Rather, pore water toxicity was highly correlated with manganese for the majority of sites sampled; subsequent spiking experiments confirmed manganese as a cause of toxicity. Current levels of manganese in the sediments of Lake Macquarie have arisen from natural sources and are not the result of anthropogenic activities. These results reiterate the importance of identifying the causes of toxicity in assessments of sediment contamination, particularly when testing sediment pore waters using sensitive early life stages.

Food concentration affects the life history response of Ceriodaphnia cf. dubia to chemicals with different mechanisms of action

The effect of three chemicals with different mechanisms of action (3,4-dichloroaniline, fenoxycarb, and chlorpyrifos) on the life history response of the cladoceran Ceriodaphnia cf. dubia was examined under both limited (3 x 10(4) cells/mL) and abundant (15 x 10(4) cells/mL) food conditions. Toxicity tests were conducted at both food concentrations simultaneously for each chemical, and cladocerans were examined daily from less than 24 h old until their death. A range of life history parameters were calculated, including mean brood sizes, survival, net reproductive rate, and population growth rate. The toxicity of 3,4-dichloroaniline was not significantly affected by food concentration. However, limited food significantly decreased the toxicity of fenoxycarb, and significantly increased the toxicity of chlorpyrifos. The effect of food concentration on toxicity appears to depend on the mechanism by which the chemical exerts its toxicity and on food--chemical interactions. Possible mechanisms for the different effects of food concentration on toxicity are discussed.

The presence of chemicals exuded by fish affects the life-history response of Ceriodaphnia cf. dubia to chemicals with different mechanisms of action

The toxicity of chemicals with different mechanisms of action (3,4-dichloroaniline, fenoxycarb, and chlorpyrifos) to the cladoceran Ceriodaphnia cf. dubia was examined in the presence and absence of chemicals exuded by fish, termed fish kairomones. A range of life-history traits were examined, including mean brood sizes, survival, net reproductive rate (Ro) and population growth rate (r). Cladocerans exposed to 3,4-dichloroaniline showed similar sensitivities in the presence and absence of fish kairomones with respect to all of the life-history traits examined. The presence of fish kairomones reduced the sensitivity of cladocerans to fenoxycarb with respect to mean brood sizes and r but increased their sensitivity in terms of Ro. The presence of fish kairomones increased the sensitivity of cladocerans to chlorpyrifos with respect to survival, r, Ro, and mean brood sizes. The general trends observed were similar to those shown when C. cf. dubia was exposed to these chemicals under low food conditions, and it is suggested that the effects of fish kairomones on toxicity may be attributed to the reduction in feeding rates observed when C. cf. dubia is exposed to fish kairomones.

Development of an improved rapid enzyme inhibition bioassay with marine and freshwater microalgae using flow cytometry

A rapid toxicity test based on inhibition of esterase activity in marine and freshwater microalgae (Selenastrum capricornutum, Chlorella sp., Dunaliella tertiolecta, Phaeodactylum tricornutum, Tetraselmis sp., Entomoneis cf. punctulata, Nitzschia cf. paleacea) was developed using flow cytometry. Uptake of fluorescein diacetate (FDA) was optimized for each species by varying the substrate concentration, incubation time, and media pH. Propidium iodide (PI) was utilized to assess membrane integrity. The optimized FDA/PI staining procedure was then used to assess the toxicity of copper in short-term exposures (1-24 h). Esterase activity was a sensitive indicator of copper toxicity in S. capricornutum and E. cf. punctulata. As copper concentrations increased, esterase activity decreased in a concentration-dependent manner. The 3- and 24-h EC50 values (based on mean activity states) were 112 microg Cu L(-1) (95% confidence limits 88-143) and 51 microg Cu L(-1) (95% confidence limits 38-70) for S. capricornutum and 47 microg Cu L(-1) (95% confidence limits 43-51) and 9.1 microg Cu L(-1) (95% confidence limits 7.6-11) for E. cf. punctulata, respectively. This enzyme inhibition endpoint showed similar sensitivity to chronic growth rate inhibition in E. cf. punctulata (48-h and 72-h EC50 values of 17 and 18 microg L(-1), respectively) but was less sensitive compared to growth for S. capricornutum (48-h and 72-h EC50 values of 4.9 and 7.5 microg L(-1), respectively). For the other five species tested, inhibition of FDA fluorescence was relatively insensitive to copper, even at copper concentrations that severely inhibited cell division rate. These short-term bioassays that detect sublethal endpoints may provide a more rapid and cost-effective way of monitoring contaminant impacts in natural waters.

Fate and toxicity of endosulfan in Namoi River water and bottom sediment

Endosulfan (6,7,8,9,10,10,-hexachloro-1,5,5a,6,9,9a-hexahydro-6,9-methano-2,4,3-benzodioxathiepine-3-oxide) sorption (standardized to 1% total organic carbon and dry weight) was significantly (P < 0.05) more concentrated on the large (>63 microm) particle fraction compared with smaller size fractions (<5 microm and 5-24 microm) of bottom sediments from the Namoi River, Australia. Following completion of the particle size fractionation (6 to 12 wk) and a sediment toxicity assessment (2 wk), the sediments showed large decreases in concentrations of alpha-endosulfan that coincided with an increase in endosulfan sulfate concentrations and minimal changes in beta-endosulfan concentrations. In the Namoi River, similar patterns were observed in the composition of total endosulfan in monthly measurements of bottom sediments and in passive samplers placed in the water column following runoff from cotton (Gossypium hirsutum L.) fields. The toxicity of endosulfan sulfate in river water indicated by the nymphs of the epibenthic mayfly Jappa kutera, was more persistent than the alpha- and beta-endosulfan parent isomers due to its longer half-life. This suggests that endosulfan sulfate would contribute most to previously observed changes in population densities of aquatic biota. Measured concentrations of total endosulfan in river water of up to 4 microg L(-1) following storm runoff, exceed the range of the 96-h median lethal concentration (LC50) values in river water for both alpha-endosulfan (LC50 = 0.7 microg L(-1); 95% confidence interval [CI] = 0.5 to 1.1) and endosulfan sulfate (LC50 = 1.2 microg L(-1); 95% CI = 0.4 to 3.3). In contrast, the 10-d LC50 value for total endosulfan in the sediment toxicity test (LC50 = 162 microg kg(-1); 95% CI = 120 to 218 microg kg(-1)) was more than threefold higher than the highest measured concentration of total endosulfan in field samples of bottom sediment (48 microg kg(-1)). This suggests that pulse exposures of endosulfan in the water column following storm runoff may be more acutely toxic to riverine biota than in contaminated bottom sediment.

Development of flow cytometry-based algal bioassays for assessing toxicity of copper in natural waters

Copper toxicity to the freshwater algae Selenastrum capricornutum and Chlorella sp. and the marine algae Phaeodactylum tricornutum and Dunaliella tertiolecta was investigated using different parameters measured by flow cytometry: cell division rate inhibition, chlorophyll a fluorescence, cell size (i.e., light-scattering), and enzyme activity. These parameters were assessed regarding their usefulness as alternative endpoints for acute (1-24 h) and chronic (48-72 h) toxicity tests. At copper concentrations of 10 micrograms/L or less, significant inhibition (50%) of the cell division rate was observed after 48- and 72-h exposures for Chlorella sp., S. capricornutum, and P. tricornutum. Bioassays based on increases in algal cell size were also sensitive for Chlorella sp. and P. tricornutum. Copper caused both chlorophyll a fluorescence stimulation (48-h EC50 of 10 +/- 1 micrograms Cu/L for P. tricornutum) and inhibition (48-h EC50 of 14 +/- 6 micrograms Cu/L for S. capricornutum). For acute toxicity over short exposure periods, esterase activity in S. capricornutum using fluorescein diacetate offered a rapid alternative (3-h EC50 of 90 +/- 40 micrograms Cu/L) to growth inhibition tests for monitoring copper toxicity in mine-impacted waters. For all the effect parameters measured, D. tertiolecta was tolerant to copper at concentrations up to its solubility limit in seawater. These results demonstrate that flow cytometry is a useful technique for toxicity testing with microalgae and provide additional information regarding the general mode of action of copper (II) to algal species.

Valve movement responses of Velesunio angasi (Bivalvia: Hyriidae) to manganese and uranium: an exception to the free ion activity model

The veracity of the free ion activity model (FIAM) was tested by examining the acute (48 h) valve movement responses (VMR) (measured in terms of the duration of valve opening) of the Australian tropical freshwater unionid bivalve, Velesunio angasi to increasing concentrations of total Mn or U, in a standard synthetic water under conditions of varying pH (5.0-6.0) and/or dissolved organic carbon (model fulvic acid, FA) concentrations (0-8.9 mg l(-1)). Valve movement behaviour, measured using an automated data acquisition system, was shown to be a quantifiable and rapid, real-time endpoint for assessing the toxic effects of Mn and U exposures. For Mn, the VMR of V. angasi were independent (P>0.05) of pH and/or model FA concentration. In contrast, VMR to U exposures were highly dependent (P< or =0.05) on pH and/or model FA concentration; individuals were more sensitive to U at low pH and model FA concentrations. Valve movement responses to Mn were directly proportional to the activity of the free metal ion (Mn(2+)), which is consistent with the FIAM. In contrast, VMR to U were regarded as an 'exception' to the FIAM, since they were a weighted function of the activities of the free metal ion and the 1:1 metal hydroxide species (i.e. 1.86 x UO2(2+) + UO2OH(+)). Additionally, the effect of U on V. angasi demonstrates the importance of examining VMR at more than one pH. At a fixed pH, the results for U were consistent with the FIAM (i.e. response was directly proportional to UO2(2+)); only when pH was altered, were the results inconsistent with the FIAM. The inconsistency in the VMR of V. angasi to U exposures in this study, together with similar examples from other studies using different metals (e.g. Al or Zn), raises questions regarding the veracity of the FIAM. A detailed examination of the conceptual development of the FIAM is required to probe its apparent failure to describe several metal-organism interactions.

pH-dependent toxicity of copper and uranium to a tropical freshwater alga (Chlorella sp.)

Copper (Cu) and uranium (U) are of potential ecotoxicological concern to tropical freshwater organisms in northern Australia as a result of mining activity. No local data on the toxicity of these metals to tropical freshwater algae are currently available. The aim of this study was to investigate the effect of pH (5.7 and 6.5) on the toxicity of Cu and U to the green alga Chlorella sp. in a synthetic softwater representative of fresh surface waters in sandy-streams of tropical northern Australia. The effects of Cu and U on algal growth (cell division) rate after a 72-h exposure were determined. Intracellular and extracellular (membrane-bound) metal concentrations at the two selected pH values were also compared. Based on the 72-h minimum detectable effect concentrations (MDEC), Chlorella sp. was approximately 20-fold more sensitive to Cu (0.7 and 1.4 µg l(-1) at pH 6.5 and 5.7, respectively) than U (13 and 34 µg l(-1) at pH 6.5 and 5.7, respectively), and more sensitive than other Australian tropical freshwater organisms. The toxicity of Cu and U was highly pH-dependent. Copper concentrations required to inhibit growth (cell division) rate by 50% (72-h EC(50)) increased from 1.5 to 35 µg l(-1) as the pH decreased from 6.5 to 5.7. Similarly, the 72-h EC(50) values for U increased from 44 to 78 µg l(-1) over the same pH range. Calculation of Cu and U speciation using the geochemical model HARPHRQ, showed that differences in the concentrations of the free metal ions (Cu(2+) and UO(2)(2+)) were only minimal (<10%) between pH 5.7 and 6.5. The decreased toxicity at pH 5.7 was due to lower concentrations of cell-bound and intracellular Cu and U compared to those at pH 6.5. These results are explained in terms of the possible mechanism of competition between H(+) and the metal ion at the cell surface.

Effect of endosulfan runoff from cotton fields on macroinvertebrates in the Namoi river

Of the several pesticides used in the pest management strategy for cotton, endosulfan is ranked as having the greatest impact on the riverine ecosystem. A survey of changes in the densities of six abundant macroinvertebrate taxa (ephemeropteran nymphs Jappa kutera, Atalophlebia australis, Tasmanocoenis sp., and Baetis sp. and two trichopteran larvae, Cheumatopsyche sp. and Ecnomus sp.) between upstream and downstream zones of the cotton-growing region in the Namoi River was conducted between November 1995 and February 1996. In November and December 1995, there were few differences in population densities between all sites. In January and February 1996, population densities of the study taxa increased 7- to 10-fold higher at the two reference sites, with low concentrations of endosulfan in sediment and in passive samplers placed in the water column. In contrast, densities of these taxa at sites with exposure to 25-fold higher concentrations of endosulfan remained static and were between one and two orders of magnitude lower than densities at the reference sites in January and February. Population densities of Baetis sp., a mobile ephemeropteran, did not indicate any inverse relationship with endosulfan concentrations. Multivariate redundancy analysis indicated that endosulfan concentrations were the leading environmental predictor of changes in density of the five benethic taxa. Laboratory 48-h LC50 values of technical endosulfan in river water were 0.6, 1.3, and 0.4 ppb for early-instar nymphs of A. australis and J. kutera, and larvae of Cheumatopsyche sp., respectively. Endosulfan sulfate formed a large proportion of the total endosulfan concentrations measured from in situ passive samplers, indicating that its main route of entry into the river is through surface runoff during storm events.

Quantitative structure-activity relationships and volume fraction analysis for nonpolar narcotic chemicals to the australian cladoceran ceriodaphnia cf. dubia

The toxicity of eleven nonpolar narcotic chemicals to the cladoceran Ceriodaphnia cf. dubia was determined. C. cf. dubia was found to be approximately four times more sensitive to these narcotic chemicals than Daphnia magna tested under virtually identical conditions. The toxicity data were also used to develop and validate quantitative structure-activity relationships (QSARs) using a range of physicochemical properties of the chemicals. The three best QSARs, based on octanol-water partition coefficients and two lipid-water partition coefficients, were able to explain 98% of the variation in toxicity. The mean absolute percentage errors between the predicted and experimental EC50 values for these three QSARs were 17.3%, 20.6%, 24.6%. Neither the critical concentration (CC) nor the critical volume (CV) hypotheses validly modeled the toxicity data when octanol-water and triolein-water partition coefficients were used although the CV hypothesis was the better of the two. When a phospholipid-water partition coefficient was used the CV hypothesis was valid. The mean toxic membrane volume fraction of 0.48 x 10(-2) m3/m3 derived in this study agreed with published values for nonpolar narcotics and supports the use of this property to determine the mode of action of chemicals.