Clinical Theriogenology 2022; 14: 44 Pregnancy and perinatal losses: general concepts Christopher Premanandan College of Veterinary Medicine, The Ohio State University, Columbus, OH Abstract Determining and, by extension, preventing loss of pregnancy is of utmost importance to a reproductive medicine practioner. How- ever, determining the cause of fetal loss can be a frustrating experience to the clinician and the diagnostician. In general, diagnostic rates are very low amongst any domestic animal species and the phrase ‘the cause of fetal loss was not apparent’ is considered a failure of the diagnostic approach. However, the owner, clinician and diagnostician should consider the value of negative findings in any abortion investigation in light of the clinical circumstances, condition of fetal tissues, diagnostic tests performed, and the gross and histologic findings. In contrast to pregnancy loss, performing postmortem diagnostics to perinatal mortality cases can be a rewarding endeavor depending on the species and the circumstances. Whereas complete postmortem examination can be intimidating to the clinician, it is a technical skill that can be easily mastered with practice. A general practitioner should be aware that portions of postmortem diagnostics (primarily the postmortem examination) can be performed ‘in house’ and may provide benefits such as timely results and increased revenue to the practice. Keywords: Abortion, fetal loss, perinatal mortality, necropsy, diagnostic pathology Introduction and definitions Pregnancy loss is roughly categorized into: 1. Embryonic death 2. Midpregnancy fetal death followed by resorption 3. Fetal death characterized by maceration, mummifica- tion or abortion 4. Stillbirth Definition of embryonic death varies among species and this variation is justified given the variable period of pregnancy in domestic animals. Species Embryonic death Equine < 42 days1 Bovine < 42 days2 Canine < 15 - 17 days Feline < 21 days Ovine < 45 days3 Caprine < 45 days4 Definition of the perinatal period varies in the veterinary lit- erature. In fact, multiple definitions are accepted in human pediatric medicine. These include infant deaths that occur pri- or to an interval from 7 to 28 days of age and fetal deaths > 20 - 28 weeks of pregnancy.5 Definitions of perinatal death in veterinary medicine are inconsistent. Some refer to deaths occurring only during midpregnancy (more traditionally re- ferred to as abortion) whereas others align more with human definitions. For the purpose of this review, perinatal death is defined as a fully formed term-animal that was delivered dead or an animal that died within 48 hours after delivery. Approach to recurrent fetal loss and perinatal loss in humans as it relates to veterinary medicine Recurrent fetal loss (RPL) is defined as the loss of > 2 preg- nancies. Since majority of pregnancy loss in humans occurs before 10 weeks, the diagnostic focus is on classifying causes of miscarriage. The classification of potential causes is well described; however, there are variations in these guidelines among international standard organizations.6 A critical fea- ture in the generation of these guidelines is the discussion of all conditions that may cause recurrent fetal loss, with recom- mendations for follow up additional testing or treatment. The value of standardizing criteria for RFL is clear. Standard- ization has generated diagnostic algorithms for clinicians to follow to advise and perhaps optimize fertility in patients. The criterion of most comparable importance in veterinary medicine is arguably the anatomic category, similar lesions in the endometrium of domestic animals. Adenomyosis, chronic endometritis and endometriosis are anatomic risk factors for RFL in humans.7 Endometriosis is only observed in primates. Adenomyosis is a common lesion of canine uterus; however, it is typically has little clinical relevance. For example, adeno- myosis was an associated finding in only 1.5% of 399 infertile bitches9 whereas, chronic inflammation is a finding associat- ed with infertility in the mare8 and the bitch.9 Clinical Theriogenology 2022; 14: 45 Efforts have been taken to understand the categories and pathogenesis of RFL (Table 1). Epigenetic modifications such as DNA methylation have a role and abnormal occurrences of DNA methylation are associated with RFL and fetal devel- opmental abnormalities.10 Inflammatory mediators such as TNF-α are implicated in RPL when there is an inappropriate shift towards Th-1 mediated inflammatory responses.11 Table 1. Commonly accepted categories associated with recurrent fetal loss Category Additional notes Cytogenetic Robertsonian translocation, trisomy Anti-phospholipid syndrome Associated with vascular thrombosis, lupus anticoagulant, anticardiolipin or antiꞴ2 glycoprotein antibody Anatomic Mullerian uterine abnormalities, uterotubal patency, uterine fibroids Hormonal or metabolic Hypothyroidism, hyperprolactinemia, Hereditary thrombocytopenia Investigation only recommended if a family history is present Male Factors DNA fragmentation, sperm aneuploidy Women with a higher incidence of miscarriage do have a high- er incidence of uterine malformations;12 however, there is lit- tle evidence that corrective surgery results in a reduction of infertility.6 Most efforts in veterinary medicine revolve around identification of infectious agents and more effort should be directed at establishing specific categories as documented with human RFL. Diagnostic modalities to consider During postmortem evaluation of adult animals, it is not un- common to recognize entities that render further diagnostic unnecessary. Examples of this include gastric dilatation volvu- lus (dogs), strangulating mesenteric lipomas (horses), and he- patic lipidosis (cats). However, pathognomonic gross lesions are uncommon in abortion cases and a uniform approach to sampling and test selection is critical to maximize diagnos- tic yield. Infectious disease is often of primary concern in the evaluation of fetal tissues and fetal membranes and perinatal losses. Detection of infectious disease relies on appropriate selection, sampling and storage of tissue during this exam- ination. A few entities causing fetal or neonatal loss can be recognized grossly. For example, canine herpesvirus-1 is of- ten associated with multifocal renal cortical hemorrhage and hemorrhage in the liver and lungs. However, bacterial septice- mia can also be associated with this lesions and polymerase chain reaction (PCR) detection of viral nucleic acid is a use- ful method to confirm the diagnosis. Similarly, cotyledonary mineralization and necrosis have been associated with ovine toxoplasmosis whereas round target-shaped regions of hepat- ic necrosis are associated with Campylobacter fetus ssp. fetus. In general, pathognomonic lesions are not common and should not be relied on exclusively for a diagnosis. Postmortem examination The technical expertise required to perform a postmortem examination should be separated from the expertise required to interpret gross tissue changes. Whereas it can take years to master interpretation of gross tissue changes, the good news is that utilizing a uniform technical approach will negate the need for meaningful gross interpretation and maximize di- agnostic yield. Equipment should be the first consideration. Regardless of species, some items will be used in common. Most of these items are related to tissue collection and storage for additional diagnostic testing. Tissue collected for histo- pathology are stored in 10% neutral buffered formaldehyde (formalin). In most circumstances, this tissue can be stored indefinitely; however, this can impede the results of immu- nohistochemical (IHC) stains. It should be noted that IHC is rarely of use in these types of cases. Clinical Theriogenology 2022; 14: 46 Figure 1. Whirl-pak bag and convenient configurations to prepare them for tissue collection Sterile collection containers are critical for fresh tissue col- lection. Whirl-pak bags are an ideal container. They are light weight, inexpensive, and easy to use. Bags should be labeled prior to starting the postmortem and as the postmortem is be- ing performed, the perforated top is torn off and the bag can be propped open with wire supports at the edge of the open- ing or by opening the base so the bag stands on its own (Fig- ure 1). Note that there are tags on the outside of the bag that are used to open the Whirl-pak bag without contaminating the inner surface. These bags should not be physically opened by inserting fingers or instruments into the inside of the bag. Prosection equipment varies depending on the species exam- ined. For small animals, a scalpel blade, Mayo scissors and forceps are typically sufficient. An older pair of scissors is preferable as they can be used to cut costochondral junction as well as sutures of the calvarium. For ruminant and equine fetuses, a sharp knife and cutters are necessary. A skilled pro- sector can perform a majority of the postmortem examination with these implements. A cleaver, handsaw or Stryker saw is typically necessary for brain removal. Culture swabs and collection of fluids is seldom helpful for microbiology and molecular diagnostics in postmortem di- agnostics. Tissues collected in a semisterile or sterile fashion is ideal. Collection of the required solid organs as soon as the thoracic and abdominal cavities are open: 1. minimizes the chance of contamination and 2. reminds the prosector to collect tissue in the first place. The only exception to this is stomach or abomasal contents. In many species, this mate- rial represents an uncontaminated source of amniotic fluid contents. This material is best stored in a small glass or plastic cylindrical container such as a red top blood tube in a sterile fashion near the end of the prosection. All containers should be sealed properly prior to shipment. Packages containing formalin jars should also contain absorbent material in case of leakage and Whirl-Pak bags should be contained in a sec- ond sealed plastic bag or container. Actual postmortem procedure varies among prosectors; how- ever, in general, are performed in either dorsal or left lateral recumbency, with the former being practical for fetal dogs and cats. Skin incisions are made from the ventral aspect of the mandible to the pubis and the skin is reflected dorsally (and bilaterally if a ventral incision is made). At this point, the abdominal wall is incised at the linea alba and additional incisions are made at the body wall just caudal to the ribcage, allowing the abdominal contents to be exposed completely (Figure 2). Depending on the approach, the thoracic cavity is opened by cutting the sternum and the dorsal aspect of the ribs in a linear fashion (with a lateral approach) or the dorsal aspect of the ribs in a linear fashion bilaterally. At this point in the postmortem, it is most prudent to col- lect the fresh tissues required for ancillary diagnostic testing. A set of clean scissors and forceps are helpful at this point. Immersing the tips of the instruments in ethanol is helpful prior to tissue collection; however, this step can be omitted if the prosector uses careful technique. Most microbiology lab- oratories will heat sterilize the outside of the tissue prior to plating the sample. Liver, lung, kidney and spleen are ideal for collection at this point and should be collected without puncturing the gastrointestinal tract. Although contamina- tion is less of an issue with a stillborn animals and aborted fetuses, this guideline should be adhered to for animals that have nursed. Stomach content collection can be reserved until the gastrointestinal tract is examined and should be stored in a separate container than the other fresh tissues. In gener- al, liver, lung, kidney, and spleen can be pooled together in 1 container for submission. Fresh tissue can be frozen until testing is performed. Tissue should be frozen at - 70˚C if being stored for long intervals but can be stored at - 20˚C if testing will take place soon after collection. Clinical Theriogenology 2022; 14: 47 Figure 2. Sampling of fresh tissue and storage of pooled tissue into a Whirl-pak immediately after opening the peritoneum and thorax Following collection of fresh specimens, tissue collection for histopathology (formalin fixation) can take place. Most diag- nostic laboratories will collect an exhaustive set of tissues for histopathology. This may be difficult to perform out in the field with equine and food animal perinatal deaths; however, it is important to remember that a very small percentage of the formalin fixed tissue is evaluated histologically (~ 3 μm tissue thickness). Large amounts of tissue are not necessary for his- tologic assessment. It is more effective to provide tissue repre- sentative of lesions and multiple small representative areas of organs. All tissue submitted in formalin should be < 0.5 cm in thickness. Eye and brain are the exceptions and can be submit- ted whole. If the organs are already < 0.5 cm in width (or close to this thickness), then they can be submitted whole. Most ca- nine or feline tissues can be submitted whole and it is unnec- essary to extensively dissect most tissues in these species (heart, kidney, etc.) aside from sectioning them to a thinner width. The order in which organs are collected is less important, although generally, the gastrointestinal tract should be reserved until af- ter solid organs have been sampled. Fetal and perinatal tissues can exhibit different physical charac- teristics than those observed in adult animals, leading to mis- interpretation of nonlesions. The most common of these is the appearance of fetal brain. Due to incomplete myelination and increased water content, the brain will be soft and difficult to handle without traumatizing the tissue. In aborted fetuses, the brain can be poured out of the open calvarium. The soft nature of the tissue does not impede histologic evaluation as long as it is adequately formalin fixed. Fetal pulmonary parenchyma tends to be reddish-purple and will often sink in formalin when obtained from stillborn animals or aborted fetuses. This is due to fetal atelectasis. If lungs consistently sink from perinates that died after parturition, this may be an abnormal finding. Fetus- es examined after an extended postmortem interval may have an accumulation of red opaque watery fluid in the peritone- al and pleural cavity. This change should generally be ignored particularly if the animal has been frozen and thawed before examination. Due to the number of misleading postmortem findings, a uniform and consistent collection of tissues in the manner outlined above should take priority over identification of potential causes during the examination. Abortion/fading neonate panels One of the more vital tools in the approach to these cases are stan- dardized panels utilized for the detection of infectious diseases. These panels employ a variety of diagnostic modalities to detect pathogens. In general, panels directed towards food animal and equine abortion/perinatal disease cases are more commonly provid- ed by state diagnostic laboratories. A panel submission brings the advantage of a simpler submission process without the need to select individual tests. However, it is a straightforward process to order the appropriate tests for a panel ‘a la carte’ that is often the case for canine and feline cases (Table 2). Table 2. Common ancillary diagnostic tests ordered in canine and feline investigations Species Tests Canine Aerobic culture, Salmonella spp. culture, Brucella canis culture, Leptospirosis spp. darkfield microscopy, Leptospirosis spp. PCR, Brucella canis PCR, Toxoplasma PCR, Canine herpesvirus-1 fluorescence antibody (FA), Canine distemper virus FA, Canine adenovirus FA Feline Aerobic culture, Feline panleukopenia PCR, Toxoplasma PCR, Feline leukemia virus indirect FA, Feline herpes PCR, Feline calicivirus FA Clinical Theriogenology 2022; 14: 48 This does require some working knowledge of testing pro- tocols and the limitations of these modalities. In either cir- cumstance, it is critical to pay attention to the requirements for tissue preservation. For instance, certain tests cannot be performed on formalin fixed tissue (culture and reverse tran- scriptase PCR). This requirement outlines the need to save a set of fresh tissue separately from the formalin fixed tissue. The most common testing modalities used include: bacteri- ologic assays, fluorescence antigen assays, viral isolation, and molecular diagnostics. 1. Bacteriologic assays In general, standard aerobic culture is acceptable in most cases. In most circumstances, tissues are preferable to swabs or fluid collection. Culture for specific pathogens that may require specific bacteriological assays include Salmonella spp. and Brucella spp.9 However, the diagnostician must keep in mind that confirmation of brucellosis is often more effectively accomplished through serologic assays. 2. Fluorescence antibody testing Also referred to as ‘direct fluorescence antibody testing,’ this test- ing modality utilizes an antibody labeled with a fluorochrome di- rected at a specific antigen associated with a pathogen. This needs to be distinguished from ‘indirect fluorescence antibody testing,’ a test that utilizes a secondary fluorochrome labeled antibody di- rected at an unconjugated antibody directed at the antigen. These methods can be used to detect a variety of pathogens in research settings but in the context of this discussion, FA is primary utilized to detect viral pathogens. Figure 3. Illustration of fluorescent antibody testing in tissue 3. Viral isolation Viral isolation is the general term used for cultivation of vi- rus particles as methods vary depending on the type of virus. Cell cultures, embryonated eggs or live animal subjects can be utilized but, in most circumstances, cell culture is used. Cell cultures can be primary (derived directly from an animal and propagated in media) or continuous immortalized cell culture lines derived from neoplasms. Either can serve as a suitable media to propagate and identify viral organisms. The type of cell line chosen for cell culture largely depends on the tropism of the virus. Cell lines of epithelial, mesenchymal and hematopoietic origin are available for this purpose. Following propagation of a virus, multiple subsequent steps can be taken to assist in identification (electron microscopy, hemagglutinin inhibition assays, complement fixation assays, and immunofluorescence assays). Viral isolation from tissue samples can be more complex that from a swab stored in vi- ral isolation media, so it is critical that as soon as possible fresh tissue be prevented from degrading. Either immediate submission for testing or freezing tissue as an intermediate step should take place. 4. Molecular diagnostics This category covers a wide variety of testing modalities but PCR is the most cited in this discussion. A wide variety of PCR modalities are available, but an exhaustive discussion is beyond the scope of this work. In general, DNA viruses are best detected using standard PCR assay whereas RNA viruses are detected using reverse transcriptase PCR. Additional mo- dalities such as multiplex PCR and nested PCR can maximize the amplification yield if there is less viral nucleic acid in the specimen. Techniques such as quantitative PCR or RT-PCR can determine the viral load in given sample but may not be critical in the diagnostic process. It is important to recall that any viral nucleic acid can be detected in PCR assay, even if in- tact viral particles are not present or are not the primary cause of the disease. Clinical Theriogenology 2022; 14: 49 Serology Whereas fetal serology can be performed, it is of little use di- agnostically unless the fetus is immunocompetent. Fetal se- rum can be collected in circumstances with mild postmortem degradation; however, it will be difficult in smaller species. Maternal serology is more convenient to obtain and more likely to provide usable diagnostic information. Paired serum samples are ideal and a necessity when certain pathogens are suspected (e.g. Leptospira spp.). The first sample is ideally col- lected at abortion and the second is collected after 2 - 3 weeks. Toxins and nutritional deficiencies Exposure to toxins is certainly associated with teratogenesis, abortion and stillbirth. Any toxin that induces severe mater- nal illness can result in secondary fetal loss. However, when referring to toxins that primarily have teratogenic effects or induce fetal loss, ruminants and horses are primarily affected. Toxins range from plant based (Ponderosa pine, locoweed), plant associated fungal products (ergot alkaloids) to pharma- ceutical agents. The ideal tissues for toxicology testing can vary depending on the toxin; however, the general tissues that ful- fill most test requirements include liver, spleen, and kidney. Conclusion Fetal and perinatal loss is a difficult scenario for most veterinarians to resolve for multiple reasons. In general, we have limited infor- mation regarding the specific causes of genetically related embry- onic loss in domestic animals.14 In ruminants, horses, and pigs, at least nutritional deficiencies, toxicities and select infectious agents are well established causes.15,16 However, more work is necessary to identify specific causes of chromosomal abnormalities and DNA damage, particularly in dogs and cats. In general, investigating fetal and perinatal loss may not be reward- ing in all cases but the technical aspect of investigating these cases can be handled in an adequate manner by most practitioners with equipment that is available in most veterinary practices. Whereas submission of a whole fetus to a diagnostic laboratory is optimal in most cases, the cost associated with the postmortem examination and ancillary diagnostic testing can be prohibitive. The option to perform the diagnostic examination ‘in house’ can provide a lower cost option to the client and increase the number of investigations in these scenarios. Conflict of interest The author has no conflict of interest. No funding was re- ceived for this publication. References 1. Vanderwall, D: Early Embryonic Loss in the Mare. J Equine Vet Sci 2008 28. 691-702. 2. Santos JE, Thatcher WW, Chebel RC, et al: The effect of embryonic death rates in cattle on the efficacy of estrus synchronization programs. Anim Reprod Sci 2004 Jul;82-83:513-35. 3. Dixon AB, Knights M, Winkler JL, et al: Patterns of late embryonic and fetal mortality and association with several factors in sheep. J Anim Sci 2007;85:1274-84. 4. Dantas AF, Riet-Correa F, Medeiros RM, et al: Embryonic death in goats caused by the ingestion of Mimosa tenuiflora. Toxicon 2012;59:555-557. 5. Barfield WD: Committee on fetus and newborn. Standard terminology for fetal, infant, and perinatal deaths. Pediatrics 2016;137. 6. Bender Atik R, Christiansen OB, Elson J, et al: ESHRE guideline: recurrent pregnancy loss. Hum Reprod Open 2018. 7. Pirtea P, Cicinelli E, De Nola R, et al: Endometrial causes of recurrent pregnancy losses endometriosis, adenomyosis, and chronic endometritis [published online ahead of print, 2021 Feb 10]. Fertil Steril 2021;S0015-0282(20)32752-7. doi:10.1016/j. fertnstert.2020.12.010. 8. Van Camp SD: Endometrial biopsy of the mare. A review and update. Vet Clin North Am Equine Pract 1988;4:229-245. 9. Gifford AT, Scarlett JM, Schlafer DH: Histopathologic findings in uterine biopsy samples from subfertile bitches: 399 cases (1990- 2005). J Am Vet Med Assoc 2014;244:180-186. 10. Pei CZ, Kim YJ, Baek KH: Pathogenetic factors involved in recurrent pregnancy loss from multiple aspects. Obstet Gynecol Sci 2019;62:212-223. 11. Saito S, Nakashima A, Shima T, et al: Th1/Th2/Th17 and regulatory T-cell paradigm in pregnancy. Am J Reprod Immunol 2010;63:601- 610. 12. Saravelos SH, Cocksedge KA, Li TC: Prevalence and diagnosis of congenital uterine anomalies in women with reproductive failure: a critical appraisal. Hum Reprod Update 2008;14:415-429. 13. Lamm CG, Njaa BL: Clinical approach to abortion, stillbirth, and neonatal death in dogs and cats. Vet Clin North Am Small Anim Pract 2012;42:501-505. 14. Khatib H, Gross N: Symposium review: Embryo survival-A genomic perspective of the other side of fertility. J Dairy Sci 2019;102:3744- 3753. 15. Mee, JF: Investigation of bovine abortion and stillbirth/perinatal mortality - similar diagnostic challenges, different approaches. Ir Vet J 2020;73:20. 16. Givens MD, Marley MS: Infectious causes of embryonic and fetal mortality. Theriogenology 2008;70:270-285.