Estimation of sensitivity and specificity of diagnostic tests and disease prevalence when the true disease state is unknown

The performance of a new diagnostic test is frequently evaluated by comparison to a perfect reference test (i.e. a gold standard). In many instances, however, a reference test is less than perfect. In this paper, we review methods for estimation of the accuracy of a diagnostic test when an imperfect reference test with known classification errors is available. Furthermore, we focus our presentation on available methods of estimation of test characteristics when the sensitivity and specificity of both tests are unknown. We present some of the available statistical methods for estimation of the accuracy of diagnostic tests when a reference test does not exist (including maximum likelihood estimation and Bayesian inference). We illustrate the application of the described methods using data from an evaluation of a nested polymerase chain reaction and microscopic examination of kidney imprints for detection of Nucleospora salmonis in rainbow trout.     

Recurrent and persistent urinary tract infections in dogs: 383 cases (1969-1995)

Laboratory records of bacterial urine cultures from 383 dogs with recurrent or persistent urinary tract infections (UTI) diagnosed at the University of California Veterinary Medical Teaching Hospital (VMTH) between 1969 and 1995 were reviewed retrospectively to characterize the bacteria involved and their association with age, gender, and breed of dogs affected. Sixty-eight breeds and a mixed-breed group were represented. Escherichia coli was the most common isolate, although mixed-bacterial infections were seen in 58% of the female and 55% of the male dogs. Recurrent and persistent UTI were most prevalent in middle-aged to older German shepherd dogs, miniature/toy poodles, and Labrador retrievers, with no apparent sex predilection. Criteria fitting recurrent and persistent UTI were present in 0.3% of all dogs seen at the VMTH during this 26-year period.     

A Polymerase Chain Reaction-Based Method to Specifically Detect Alternaria alternata Apple Pathotype (A. mali), the Causal Agent of Alternaria Blotch of Apple

ABSTRACT Alternaria alternata apple pathotype (previously A. mali) causes Alternaria blotch on susceptible apple cultivars through the production of a host-specific toxin, AM-toxin. Identification of some Alternaria species, especially those that produce host-specific toxins, has been extremely difficult due to a high level of variability which extends even to nonpathogenic isolates. We have recently cloned and characterized a gene (AMT) that plays a crucial role in AM-toxin biosynthesis and demonstrated that it is only present in isolates of A. alternata apple pathotype. Using primers designed for the AMT gene, we developed a polymerase chainreaction-based method to specifically detect AM-toxin producing isolates of A. alternata apple pathotype.     

Dystocia in an injured common eastern long-necked turtle (Chelodina longicollis).

Dystocia in the class Reptilia is a common problem. In turtles, it is often difficult to distinguish between a normal gravid state and dystocia. This case report describes nonobstructive dystocia in a free-living freshwater chelonian, Chelodina longicollis, complicated by traumatic injuries to the head, bridge, and plastron. Medical treatment of the dystocia and external fixation of the facial injuries provided a successful outcome.     

Systemic blastomycosis in a horse.

Progressive multisystemic disease caused by Blastomyces dermatitidis was diagnosed in a 17-year-old Quarter horse broodmare. The mare had been treated unsuccessfully with antibiotics for mastitis 3 months postpartum. The disease progressed to exudative cutaneous lesions affecting the ventrum, pectoral region, and limbs accompanied by weight loss across several months. Yeast bodies were observed in swabs of the cutaneous exudate, suggesting a clinical diagnosis of blastomycosis. Following referral, pleural effusion, cavitated lung lesions, and hyperproteinemia were identified, and the mare was euthanized because of poor prognosis. Necropsy revealed extensive pyogranulomas in the mammary gland, skin, subcutaneous tissues, and lungs, accompanied by thrombi in major blood vessels of the lungs and hind limbs. Histologically, pyogranulomatous inflammation was evident in many tissues, and fungal organisms were seen in sections of mammary gland, skin, subcutis, pericardium, and lung. Blastomyces dermatitidis was cultured from mammary tissue, lungs, lymph node, and an inguinal abscess. Although blastomycosis is endemic in the area of origin of the mare in northwestern Wisconsin, the disease is extremely rare in horses and hence easily misdiagnosed. Unique features of this case included the extent of mammary gland involvement and the presence of thrombi in multiple sites.     

Development of an assay to determine single nucleotide polymorphisms in the prion gene for the genetic diagnosis of relative susceptibility to classical scrapie in sheep.

The objective of this study was to develop a reliable Taqman 5' Nuclease Assay for genotyping sheep for scrapie susceptibility. The sheep prion gene contains 2 single nucleotide polymorphisms (SNPs) that may mediate resistance to classical scrapie, one at codon 136, alanine (A) or valine (V), and another at codon 171, arginine (R) or glutamine (Q). The R allele appears to confer resistance to classical scrapie, with the AA(136) RR(171) genotype the most resistant to scrapie and QR(171) only rarely infected in the US sheep population. The Assays by Design protocol was used for development of probes and primers for codon 136 and Primer Express for codon 171. Commercially available kits were used to isolate genomic DNA from blood or muscle. For validation, 70 SNP determinations for each codon were compared to commercial testing with an error rate of less than 1%. Then, 935 samples from blood (n = 818) and muscle (n = 117) were tested for both codons with 928 successful determinations and only 7 samples (<1% of total samples) that needed repeating. Genotypes were AA QQ (n = 102; 11.0%), AV QQ (n = 28; 3.0%), AA QR (n = 396; 42.7%), AV QR (n = 54; 5.8%), and AA RR (n = 348; 37.5%). Thus, 86% of the sheep tested (n = 798) contained R at codon 171 and were expected to be scrapie-resistant. This new Taqman 5' Nuclease SNP genotyping assay is accurate, easy to perform, and useful in the study of classical scrapie in sheep and its prevention through selective breeding programs to eliminate highly susceptible animals.     

Geospatial methods provide timely and comprehensive urban forest information

Urban forests are unique and highly valued resources. However, trees in urban forests are often under greater stress than those in rural or undeveloped areas due to soil compaction, restricted growing spaces, high temperatures, and exposure to air and water pollution. In addition, conditions change more quickly in urban as opposed to rural and undeveloped settings. Subsequently, proactive management of urban forests can be challenging and requires the availability of current and comprehensive information. Geospatial tools, such as, geographic information systems (GIS), global positioning systems (GPS) and remote sensing, work extremely well together for gathering, analyzing, and reporting information. Many urban forest management questions could be quickly and effectively addressed using geospatial methods and tools. The geospatial tools can provide timely and extensive spatial data from which urban forest attributes can be derived, such as land cover, forest structure, species composition and condition, heat island effects, and carbon storage. Emerging geospatial tools that could be adapted for urban forest applications include data fusion, virtual reality, three-dimensional visualization, Internet delivery, modeling, and emergency response.     

Results of grid keratotomy, superficial keratectomy and debridement for the management of persistent corneal erosions in 92 dogs

Ninety-two cases of persistent corneal erosions in dogs were analyzed for breed, gender, age and which eye was affected. The results of the treatment of 92 persistent corneal erosions in dogs by superficial keratectomy (SK), grid keratotomy (GK), or debridement with a sterile dry cotton swab are presented. These techniques gave better rates of healing than have been previously reported. All cases of persistent corneal erosions healed in this study. However, it must be noted that three cases treated with debridement only failed to heal after several treatments and were eventually treated with SK. After one procedure 80 out of 92 (87%) had healed. After one procedure, 63% of cases treated with debridement healed, 100% of cases treated with SK healed, and 85% of cases treated with GK healed. At the first postoperative visit, 88% (21/24 cases) of ulcers treated by SK had healed, and 75% (39/52 cases) of ulcers treated by GK had healed. Only 25% of the persistent corneal erosions had healed at the first visit after debridement. All 24 cases of persistent corneal erosions treated with SK healed after one treatment in a mean +/- SD of 9.3 +/- 3.9 days (median of 7 days). Fifty-two cases were managed with GK; 44 (83%) of these healed with one procedure and eight cases required a second GK procedure to resolve. A mean +/- SD of 13.4 +/- 5.1 days (median of 11.5 days) following GK was required for the persistent corneal erosions to heal. Nineteen cases were initially managed by debridement with a dry cotton swab under local anesthesia. Sixteen out of these 19 debridement cases healed (giving an overall healing rate of 84%) in a mean +/- SD time of 23.4 +/- 11.1 days (median 21.5). There were three cases that did not heal with debridement. These cases were debrided at 10-20 day intervals for 30-60 days, and were then treated with SK. Two of these cases healed within 7 days, the other case required 18 days to heal. Sixty-three per cent of persistent corneal erosions treated with debridement healed after one procedure; however, only four out of 19 cases (21%) were healed at the first revisit. Complications were rare: corneal edema occurred in two cases following multiple GK, and excessive granulation tissue in one case was managed with a SK. There was the occurrence of an ulcer adjacent to the surgery site in four cases, two cases following GK and two cases following SK.     

Failure of passive transfer in foals: incidence and outcome on four studs in New South Wales

Objective To determine the regional incidence and effectiveness of treatment of failure of passive transfer (FPT) in foals. Design A study of disease incidence. Animals Eighty-eight foals and 57 mares from four studs in the practice area of the Rural Veterinary Centre were tested. Procedure Foals were tested for their serum IgG and total serum protein (TSP) concentration within the first 72 hours of life. Colostrum was collected from mares and specific gravity determined. FPT and partial failure of passive transfer (PFPT) of immunoglobulins was diagnosed when serum IgG concentrations were < 4 g/L and 4 to 8 g/L respectively. Owners of foals diagnosed with FPT were offered treatment with 1 to 2 L plasma (TSP > 70 g/L); 9 (64%) of the affected foals were treated. Results Fourteen foals (16%) had FPT whereas 15 (17%) had PFPT. There were significant differences between the mean TSP concentration in foals with FPT (42.6 ± 4.2 g/L), PFPT (48.1 ± 3.9 g/L) and those acquiring adequate passive immunity (58.9 ± 5.5 g/L) (P < 0.01). Sixteen (29%) mares had pre-suck colostral specific gravity < 1.060 and 12 (71%) foals raised by these mares had FPT or PFPT. The incidence of severe disease (categorised by a sepsis score > 11, positive culture of bacteria from blood or disease requiring hospitalisation) in all foals in the first 2 months of life was 10%. However, none of the nine foals with FPT that received plasma experienced severe disease. In contrast, foals with PFPT had an increased susceptibility to severe disease (P < 0.001) when compared with normal foals. Conclusion Treatment of foals with FPT may reduce the subsequent incidence of severe disease. Pre-suck colostral specific gravity and foal TSP may be used to predict the likelihood of FPT and PFPT. Even though the number of foals studied is small the results highlight the importance of optimal management practices in reducing the incidence of FPT and disease associated with this process.