Abstract
Neonatal mortality rate (NMR) may be affected by maternal physical condition, anesthesia, and uterine incision or en-bloc neonate removal. The association of selected factors with number of dogs with dead puppies at hospital discharge was evaluated using 78 records. Data obtained at admittance for emergency cesarean section included: age, small or large body size, rectal temperature, packed cell volume, serum total protein, blood urea nitrogen, glucose, puppy in pelvic canal, and heart rate. Administration of opioids, propofol, alfaxalone, isoflurane, and sevoflurane, and anesthesia/surgical times and surgical technique were evaluated using Fisher’s exact tests. There were 238 live puppies and 38 dogs had 58 dead puppies; the NMR was 19.6%. Mortality was associated with puppy in the pelvic canal (P = 0.003) and duration of anesthesia > 80 minutes (P = 0.029). Age > 8 years (P = 0.054) and induction time to start of surgery > 30 minutes (P = 0.17) may be associated with mortality. Expedient cesarean section with obstructive dystocia and an induction time to start of surgery < 30 minutes are important for puppy survival.
Résumé
Influence des facteurs maternels, anesthésiques et chirurgicaux sur la survie néonatale après césarienne d’urgence chez 78 chiens : Une étude rétrospective (2002 à 2020). Le taux de mortalité néonatale (TMN) peut être affecté par l’état physique de la mère, l’anesthésie et l’incision utérine ou l’ablation en bloc du nouveau-né. L’association de facteurs sélectionnés avec le nombre de chiens avec des chiots morts à la sortie de l’hôpital a été évaluée à l’aide de 78 dossiers. Les données obtenues à l’admission pour une césarienne d’urgence comprenaient : l’âge, la taille corporelle petite ou grande, la température rectale, l’hématocrite, les protéines sériques totales, l’azotémie, le glucose, la présence de chiot dans le canal pelvien et la fréquence cardiaque. L’administration d’opioïdes, de propofol, d’alfaxalone, d’isoflurane et de sévoflurane, ainsi que les temps d’anesthésie/de chirurgie et la technique chirurgicale ont été évalués à l’aide des tests exacts de Fisher. Il y avait 238 chiots vivants et 38 chiens avaient 58 chiots morts; le TMN était de 19,6 %. La mortalité était associée la présence de chiot dans le canal pelvien (P = 0,003) et à la durée de l’anesthésie > 80 minutes (P = 0,029). Un âge > 8 ans (P = 0,054) et un délai d’induction avant le début de la chirurgie > 30 minutes (P = 0,17) peuvent être associés à la mortalité. Une césarienne opportune lors de dystocie obstructive et un temps d’induction avant le début de la chirurgie < 30 minutes sont importants pour la survie du chiot.
(Traduit par Dr Serge Messier)
Introduction
Several maternal and neonatal factors can influence neonatal survival during parturition and are reviewed elsewhere (1–4). Various maternal and fetal causes of dystocia may require cesarean section to ensure maternal and neonatal survival, and an emergency is a recognized risk factor (5,6). Dogs with dystocia can be dehydrated, hypovolemic, hypotensive, and toxemic as problems progress without intervention. The neonatal survival rate for emergency cesarean section is a third of that for elective cesarean section (5) and dogs are 3 times more likely to have dead puppies if they have emergency cesarean section compared to those that have elective cesarean section (6). In a prospective multicenter study evaluating 3908 cesarean section-derived puppies, the neonatal mortality rate (NMR) was 8% immediately after cesarean section, 13% 2 h after removal, and 20% 1 wk later (5). In contrast, the NMR for naturally born neonates is 2.2% at birth and 8% after 24 h (7).
Sedative and anesthetic drugs are lipophilic, ensuring rapid transfer from the circulation to the sites of action in the central nervous system; however, their lipophilic nature also enables rapid transfer across the placenta to the fetus (1). The fetus and neonate are affected by drug actions but have reduced capacity to metabolize certain drugs (1,8,9). Drugs with rapid elimination kinetics are recommended for general anesthesia of the dam (1,8–10). Induction with an injectable drug with rapid distribution and redistribution phases allows a concentration gradient across the placenta that promotes drug transfer from the fetus back to the dam for metabolism or elimination (1,2). At least 15 min from injectable induction of anesthesia to removal of the puppy has been suggested before the fetus is removed from the placenta (10,11). A delivery time < 10 min did not result in increased neonatal survival (5) and neonate Apgar scores improved 30 min following induction of anesthesia with dexmedetomidine and propofol (12). Increased neonatal blood-brain barrier permeability results in a greater sensitivity to opioid drug effects compared to adults (8,9), but opioids are used for maternal sedation and analgesia because effects can be reversed in the neonate with naloxone (1). Xylazine has been associated with increased NMR, but medetomidine appears to not affect NMR if atipamezole is administered to the neonate (4,13).
For cesarean section, suggested options for general anesthesia include minimal sedation and induction with rapidly eliminated injectable agents followed by maintenance with isoflurane or sevoflurane. Facemask induction with an inhalational agent after prior sedation with an opioid was used in the past when rapidly eliminated injectable induction drugs were not available (1,5,10). Recently, induction of anesthesia with propofol or alfaxalone with or without prior sedation has become standard practice (1,5,10–18) and both drugs allow delivery of vigorous puppies when isoflurane is administered for maintenance (14,15). One study reported administration of alfaxalone resulted in more vigorous puppies in the first 60 min following removal compared to administration of propofol (16). Isoflurane or sevoflurane anesthesia is recommended for maintenance of anesthesia because of rapid pulmonary elimination with minimal dependence on metabolism (17). Puppy vitality was greater with isoflurane compared to alfaxalone intravenous infusion, although neonatal survival was similar (18). Modern rapidly eliminated drugs are not linked with increased NMR (5,17).
Either uterine incisions or en-bloc removal of the uterus with live fetuses is used to extract puppies, and both techniques have a similar NMR (19,20). Uterine incisional technique can delay removal of all puppies, if there are many present, prolonging exposure of the fetus to anesthetic gases and reduced placental perfusion, but the en-bloc method can increase NMR from fetal hypoxia if neonates are not removed within 60 s of arterial clamping and immediately resuscitated (4,19).
The objective of this retrospective study was to determine if there is an association between certain maternal factors, timing of anesthesia and surgery, drugs used for anesthesia, or surgical technique with neonatal mortality in dogs undergoing cesarean section. The null hypothesis was that maternal factors, choice of drugs, anesthetic and surgical times, and surgical technique would have no effects on neonatal mortality.
Materials and methods
Records were obtained from a veterinary teaching hospital incorporating primary care and referral practices. The clinic operates an emergency service with interns under supervision of small animal medicine internists for its own primary care practice and for neighboring primary care veterinary clinics. Specialist surgeons (residents/Boarded surgeons) performed cesarean sections with a dedicated anesthesia service directly provided by students or technicians who are closely supervised by anesthesia residents or Boarded anesthetists. The institutional medical records from 2002 to 2020 were searched for dogs undergoing emergency cesarean section using retrieval terms dystocia; C section; and cesarean/caesarean section for dogs. Exclusion criteria included: elective cesarean section, grossly incomplete anesthesia, and medical records, records with unresolvable discrepancies, and known dead fetuses prior to cesarean section. In records that were included, puppies born naturally before cesarean section, severe malformation, mummified or autolyzed fetuses were not counted.
Variables of interest at time of admittance into the hospital included: age (< 8 y and ≥ 8 y), small (< 10 kg) or large (> 10 kg) dog, packed cell volume (PCV), total blood solids (TS), blood urea nitrogen (BUN), blood glucose, rectal body temperature (BT), and heart rate. The American Society of Anesthesia (ASA) physical status was divided into 2 categories of ASA 1–2 and ASA 3–5 for comparison. The PCV and TS were measured using a microhematocrit and refractometer. The glucose and BUN were measured in blood using commercial reagent sticks (AlphaTRAK 2; Zoetis Canada, Kirkland, Quebec, Azostix; Siemens Healthcare Diagnostics, Tarrytown, New York, USA, respectively).
Drugs written on the anesthesia record under “Premedication” were considered as “sedation.” Drugs used to induce anesthesia for endotracheal intubation were written on the record under “Induction” and recorded. A facemask was used to administer inhalational agents for induction of anesthesia where indicated. On the record, a checked box indicated whether the inhalational agent for maintenance of anesthesia was isoflurane (AErrane; Baxter, Mississauga, Ontario) or sevoflurane (Sevorane; AbbVie, Montreal, Quebec). Whether the procedure was performed during the day (8 am to 6 pm) or night (6 pm to 8 am), time from induction of anesthesia to start of surgery and total anesthesia and surgery times were recorded.
The surgical report described either a conventional uterine incision for removal of neonates or en-bloc removal of the uterus. Dogs with neonates removed by uterine incision and spayed afterwards were included under the conventional surgical technique category. The number of puppies present in the uterus and pelvic canal, and the number discharged from the clinic with the dam were recorded. The number of dams with 1 or more dead puppies at time of discharge was recorded and the numbers of dead and live puppies were used to calculate the NMR.
Generally, the approach to surgery was to aseptically insert IV catheters and provide all dogs with IV antimicrobial and crystalloid fluid therapy [5 to 10 mL/kg body weight (BW) per hour] with increased rates when required, based on history, physical examination, and emergency blood analysis. Calcium gluconate and IV dextrose or oral corn syrup were administered before anesthesia if the ionized calcium or blood glucose was below normal, respectively. Preoxygenation was provided either by facemask or “flow-by” technique depending on patient compliance. Minimum anesthetic monitoring consisted of assessment of depth using eye position, jaw tone, and palpebral reflexes, indirect blood pressure measurement, pulse oximetry, capnometry, and volatile gas measurement. Hypotension (mean arterial blood pressure ≤ 60 mmHg) was treated with crystalloid and colloid fluid bolus, ephedrine IV bolus, and dobutamine IV infusion as dictated by the primary anesthetist. Intermittent positive pressure ventilation was used when deemed necessary by the attending anesthetist. Neonatal resuscitation generally included stimulation, application of warmth, oxygen, administration of doxapram and/or naloxone, tracheal intubation, and lung ventilation when necessary.
Statistics
Descriptive statistics were used to provide mean ± standard deviation (SD) for maternal body weight, age, rectal temperature, heart rate, PCV, TS, blood glucose, BUN, and selected dose rates. A Shapiro-Wilk test was used to assess normality. Time from induction to start of surgery was compared among dogs with dead puppies and dogs with surviving puppies using an unpaired t-test. Total surgery and anesthesia times were compared using a Mann-Whitney U-test.
The temperature, heart rate, and blood analysis data were divided into normal and abnormal data using available reference ranges (21,22). The number of dogs with 1 or more dead puppies or all live puppies were evaluated with drugs used, timing of procedure, anesthesia and surgery times, presence of puppy in pelvic canal, surgical technique, and normal and abnormal categories using 2 × 2 contingency tables and a Fisher’s exact test. Significance was set at P < 0.05. Odds ratios and 95% confidence intervals (CI) were calculated using the Baptista-Pike method to evaluate categorial influence on risk of a dog having one or more dead puppies. Analyses were performed using GraphPad Prism version 9.0.0 for Windows (GraphPad Software, San Diego, California, USA).
Results
A total of 78 institutional medical records of cesarean section were eligible for enrollment and no maternal deaths were recorded. A total of 38 dogs had 1 or more dead puppies by the time of discharge from the hospital. The total number of puppies delivered was 296 and total number of neonatal deaths before discharge was 58. The NMR was 19.6%. There were 19 dogs out of a total of 37 dogs with dead puppies from 2002 to 2010 (51.4%) and 19 dogs out of 41 dogs from 2011 to 2020 (46.3%) with no significant difference between these 2 time periods (P = 0.821). From 20 complete records, the time the dam was hospitalized was 11.6 ± 8.4 h (range: 5 to 39 h). Two dogs had the largest litter size of 10 puppies and 1 of these dogs had a dead puppy. Overall mean ± SD and range of maternal body weight before cesarean section was 17.9 ± 15.5 kg (range: 2.3 to 64.0 kg). There were 38 small dogs and 40 large dogs. The mean ± SD and range of age was 4.0 ± 2.1 (1 to 10) y. There were 71 dogs < 8 y old and 7 dogs ≥ 8 y. Further details are presented in Tables 1 to 5.
Table 1.
Breed description of 78 dogs enrolled in a retrospective study evaluating the neonatal mortality rate over an 18-year period at a combined primary and referral care practice. Crossbred dogs have been labelled with the closest breed description.
| Breed description | Frequency | Number of dogs with dead puppies |
|---|---|---|
| Chihuahua | 8 | 2 |
| Yorkshire terrier | 8 | 4 |
| Shih tzu | 7 | 3 |
| Pug | 6 | 3 |
| German shepherd dog | 6 | 4 |
| Labrador retriever | 5 | 2 |
| Golden retriever | 2 | 1 |
| Havanese | 2 | 0 |
| Scottish terrier | 2 | 1 |
| Siberian husky | 2 | 2 |
| Staffordshire terrier | 2 | 1 |
| Jack Russell terrier | 2 | 2 |
| Newfoundland | 2 | 1 |
| Toy poodle | 2 | 0 |
| American pitbull, basset hound, Belgian Malinois shepherd, blue heeler, Boston terrier, boxer, border collie, bulldog, cocker spaniel, elkhound, English springer spaniel, flat-coated retriever, French bulldog, German short-haired pointer, Large Munsterlander, Maltese, old English sheepdog, Papillon, Pomeranian, Rhodesian ridgeback, rough collie, whippet | 1 of each | 12 |
Table 2.
Details from 78 dogs enrolled into a retrospective study evaluating neonatal mortality over an 18-year period.
| Variable | Mean ± standard deviation (range) | Ranges used to define abnormal | Dogs with dead puppies/total dogs | P-value | Odds ratio (95% CI) |
|---|---|---|---|---|---|
| Age < 8 y | 3.6 ± 1.7 (1 to 6) | na | 32/71 | 0.054 | 0.1 |
| Age ≥ 8 y | 8.4 ± 0.9 (8 to 10) | 6/7 | (0.01 to 1.0) | ||
| Body weight ≤ 10 kg | 5 ± 2.2 (2.3 to 9.2) | na | 16/38 | 0.267 | 0.6 |
| Body weight > 10 kg | 30 ± 12.4 (12.2 to 64.0) | 22/40 | (0.3 to 1.5) | ||
| ASA status 1–2 | na | na | 16/32 | 0.802 | 1.2 |
| ASA status 3–5 | 13/28 | (0.4 to 3.0) | |||
| PCV (%) | 40.4 ± 6.7 (25 to 61) | Normal | 28/65 | 0.509 | 2.0 |
| < 35 or > 55% | 3/11 | (0.5 to 7.5) | |||
| TS (g/L) | 67.8 ± 8.1 (36.0 to 87.0) | Normal | 26/58 | > 0.999 | 1.0 |
| < 55 or > 73 g/L | 8/18 | (0.4 to 3.2) | |||
| BUN 1.8 to 5.4 mmol/L | na | Normal | 17/42 | 0.181 | 0.5 |
| BUN 5.4 to 9.3 mmol/L (n = 16) | Abnormal | 12/20 | (0.2 to 1.3) | ||
| BUN 9.3 to 14.3 mmol/L (n = 1) | |||||
| Blood glucose (mmoL/L) | 5.8 ± 1.7 (2.1 to 9.5) | Normal | 23/44 | 0.453 | 1.6 |
| < 3.6 or > 6.7 mmoL/L | 10/25 | (0.6 to 4.3) | |||
| Rectal temperature (°C) | 38.0 ± 0.7 (35.4 to 39.8) | Normal | 28/59 | 0.786 | 0.8 |
| < 37.5 or ≥ 38.9°C | 9/17 | (0.3 to 4.3) | |||
| Heart rate (beats/min) | 137 ± 26 (72 to 200) | Normal | 31/65 | 0.733 | 0.7 |
| ≥ 180 beats/min | 5/9 | (0.2 to 2.8) | |||
| All puppies in uterus only | na | na | 26/64 | 0.003* | 0.1 |
| Dogs with a puppy in pelvic canal | 12/14 | (0.02 to 0.6) | |||
| Procedure performed during day | na | na | 14/29 | 0.796 | 1.2 |
| Procedure performed during night | 13/30 | (0.5 to 3.2) | |||
| Induction-surgery start ≤ 15 min | na | na | 6/18 | 0.170 | 0.4 |
| Induction-surgery start > 15 min | 22/41 | (0.1 to 1.4) | |||
| Induction-surgery start ≥ 30 min | na | na | 7/10 | 0.169 | 3.1 |
| Induction-surgery start < 30 min | 21/49 | (0.7 to 11.9) | |||
| Anesthesia time ≤ 80 min | na | na | 9/28 | 0.029* | 0.3 |
| Anesthesia time > 80 min | 24/40 | (0.1 to 0.9) | |||
| Surgery time ≤ 50 min | na | na | 4/13 | 0.327 | 0.4 |
| Surgery time > 50 min | 16/32 | (0.1 to 1.7) |
n — Indicates number of dogs for data analysis; na — Not applicable; CI — Confidence interval.
Significantly different between categories.
PCV — packed cell volume, TS — total solids, BUN — blood urea nitrogen.
Dogs were categorized by age, body size, and American Society of Anesthesia (ASA) physical status 1–2 or 3–5, whether a puppy was present in the pelvic canal, and when the procedure was performed. The number of dogs with 1 or more dead puppies are presented for normal and abnormal hematology, biochemistry, and physical factors with total dogs. Comparisons were made between categories using 2 × 2 contingency tables and a Fisher’s exact test. P-value was set at 0.05. Odds ratio is risk of dogs having 1 or more dead puppies in young, small, ASA 1–2, no puppy in pelvic canal, ≤ 15 or ≥ 30 min from induction to start of surgery, short anesthesia and surgery times, day surgery or normal categories (< 1 is a low risk; > 1 is a high risk).
Table 3.
Details of comparisons made between administered anesthetic drugs and surgical techniques for 78 dogs enrolled in a retrospective study evaluating neonatal mortality over an 18-year period.
| Category comparisons | Total number of dogs in category | Number of dogs with dead puppies | P-value | Odds ratio (95% CI) |
|---|---|---|---|---|
| Dogs induced with facemask inhalant | 3 | 3 | 0.111 | Infinity |
| Dogs induced with propofol and alfaxalone | 75 | 35 | (0.9 to Infinity) | |
| Dogs maintained with isoflurane | 36 | 18 | > 0.999 | 1.1 |
| Dogs maintained with sevoflurane | 38 | 18 | (0.5 to 2.7) | |
| Dogs induced with alfaxalone with or without opioid | 22 | 8 | 0.313 | 0.6 |
| Dogs induced with propofol with or without opioid | 53 | 27 | (0.2 to 1.6) | |
| Dogs with opioid sedation and induced with propofol | 12 | 8 | 0.327 | 2.3 |
| Dogs with no opioid sedation and induced with propofol | 41 | 19 | (0.6 to 7.7) | |
| Dogs with hydromorphone sedation with propofol induction | 3 | 1 | 0.346 | 3.5 |
| Dogs induced with propofol without opioid sedation | 41 | 19 | (0.5 to 46.9) | |
| Dogs with fentanyl sedation and propofol induction | 6 | 3 | > 0.999 | 1.2 |
| Dogs induced with propofol without opioid sedation | 41 | 19 | (0.3 to 5.4) | |
| Dogs sedated with remifentanil with propofol induction | 1 | 1 | 0.476 | Infinity |
| Dogs induced with propofol without opioid sedation | 41 | 19 | (0.1 to Infinity) | |
| Dogs sedated with sufentanil with propofol induction | 1 | 1 | 0.476 | Infinity |
| Dogs induced with propofol without opioid sedation | 41 | 19 | (0.1 to Infinity) | |
| Dogs with all opioid sedation with both injectable induction | 25 | 14 | 0.328 | 1.8 |
| Dogs without opioid sedation with both injectable induction | 50 | 21 | (0.7 to 4.4) | |
| Dogs with hydromorphone sedation with both injectable induction | 5 | 3 | 0.642 | 2.1 |
| Dogs without opioid sedation with both injectable induction | 50 | 21 | (0.4 to 12.7) | |
| Dogs with fentanyl sedation with both injectable induction | 15 | 7 | 0.774 | 1.2 |
| Dogs without opioid sedation with both injectable induction | 50 | 21 | (0.4 to 3.5) | |
| Dogs with remifentanil sedation with both injectable induction | 4 | 3 | 0.312 | 4.1 |
| Dogs without opioid sedation with both injectable induction | 50 | 21 | (0.6 to 55.4) | |
| Dogs with sufentanil sedation with propofol (not inhalant) | 1 | 1 | 0.431 | Infinity |
| Dogs without opioid sedation with both injectable induction | 50 | 21 | (0.2 to Infinity) | |
| Dogs with fentanyl sedation with alfaxalone induction | 9 | 4 | 0.62 | 2.8 |
| Dogs with alfaxalone with no fentanyl sedation | 9 | 2 | (0.4 to 18.2) | |
| Dogs with uterine incisional cesarean section including dogs spayed afterwards | 61 | 29 | 0.765 | 0.8 |
| Dogs with en-bloc uterus removal with neonates | 13 | 7 | (0.2 to 2.5) |
CI — Confidence interval.
Comparison made between the top listed category and the category listed immediately below, using 2 × 2 contingency tables and a Fisher’s exact test. P-value was set at 0.05. Odds ratio is risk of having one or more dead puppies in top category compared to category beneath (< 1 is a low risk; > 1 is a high risk).
Table 4.
Time from induction of anesthesia to start of surgery, total anesthesia, and surgical duration in 78 dogs undergoing emergency cesarean section at a combined primary care and referral institution.
| Period | All dogs | Dogs with only live puppies | Dogs with 1 or more dead puppies | P-value |
|---|---|---|---|---|
| Induction of anesthesia to start of surgery (min) | 22 ± 9 | 20 ± 8 | 24 ± 9 | 0.039* |
| (7 to 45) | (7 to 38) | (8 to 45) | ||
| n = 59 | n = 31 | n = 28 | ||
| Total anesthesia time (min) | 92 ± 33 | 86 ± 35 | 98 ± 29 | 0.043* |
| (30 to 215) | (40 to 215 | (30 to 165) | ||
| n = 68 | n = 35 | n = 33 | ||
| Total surgical time (min) | 66 ± 27 | 65 ± 29 | 68 ± 25 | 0.043* |
| (22 to 159) | (25 to 159) | (22 to 136) | ||
| n = 45 | n = 25 | n = 20 |
n — Represents number of dogs in that category from available records.
Represents statistical significance (P < 0.05) in time between dogs with only live puppies and dogs with one or more dead puppies at time of hospital discharge. Data presented as mean ± standard deviation with range in brackets.
Table 5.
Intravenous doses of injectable induction drugs used in 78 dogs anesthetized for emergency cesarean section at a combined primary care and referral institution.
| Induction drug | Dose when drug given alone | Dose after opioid sedation |
|---|---|---|
| Propofol (mg/kg BW) | 5.4 ± 1.6 | 4.8 ± 1.8 |
| (0.7 to 10) | (1.9 to 8.0) | |
| n = 41 | n = 12 | |
| Alfaxalone (mg/kg BW) | 2.5 ± 0.8 | 2.3 ± 1.1 |
| (1.7 to 4.0) | (1.3 to 5.6) | |
| n = 9 | n = 13 |
Data are presented as mean ± standard deviation with range in brackets. BW — body weight.
n — Represents number of dogs in that category from available records.
One dog with a PCV of 28% and glucose concentration of 2.4 mmoL/L had a blood transfusion before anesthesia and no puppies died, but the reason for anemia was not recorded. One dog on admission had a low blood glucose level of 2.1 mmoL/L and was hypocalcemic with 1 dead puppy from a litter of 6. Three other dogs were treated with calcium gluconate and 1 dog had a dead puppy. One dog had a BUN in the range 10.7 to 14.3 mmoL/L and had 2 dead and 3 live puppies. Six out of eight hyperthermic dogs had dead puppies.
All sedative drugs were opioids and included hydromorphone (Hydromorphone, 2 or 10 mg/mL; Sandoz Canada, Boucherville, Quebec), 0.05 to 0.1 mg/kg BW, IV, fentanyl (Fentanyl citrate 0.05 mg/mL; Sandoz Canada), 0.002 to 0.005 mg/kg BW, IV, remifentanil (Remifentanil hydrochloride for injection; SteriMax, Oakville, Ontario), 0.002 to 0.005 mg/kg BW, IV, or sufentanil (Sufentanil citrate 0.05 mg/mL; Sandoz Canada), 0.0025 mg/kg BW, IV, administered 5 to 10 min before induction of anesthesia. Three dogs (4%) were induced with sevoflurane by facemask, 1 without prior sedation, and 2 were sedated with sufentanil IV before induction; all 3 dogs had dead puppies. Three other dogs (4%) were administered propofol (Diprivan; AstraZeneca Canada, Mississauga, Ontario), an average of 4.9 mg/kg BW, IV and induction was completed in 2 dogs with isoflurane and the third dog with sevoflurane; 1 dog that was given isoflurane had dead puppies.
A total of 41 dogs (52.6%) were induced with IV propofol without prior sedation; 19 dogs had dead puppies. Twelve dogs (15.4%) were induced with IV propofol after opioid sedation; 8 dogs had dead puppies. Of these 12 dogs, 4 dogs were sedated with hydromorphone (3 dogs had dead puppies); 6 dogs were sedated with fentanyl (3 dogs had dead puppies); 1 dog was sedated with remifentanil and 1 dog was sedated with sufentanil and both had dead puppies.
Twenty-two dogs (28.2%) were induced with IV alfaxalone (Alfaxan 10 mg/mL; Jurox Pty, NSW, Australia). Nine dogs were induced with alfaxalone without prior sedation and 2 dogs had dead puppies. Three dogs were induced with alfaxalone after sedation with remifentanil and 2 dogs had dead puppies. One dog was sedated with hydromorphone with live puppies. Nine dogs were sedated with fentanyl and induced with alfaxalone and 4 dogs had dead puppies.
For maintenance of anesthesia, 36 dogs (46.2%) were administered isoflurane and 18 dogs had dead puppies. Sevoflurane was administered to 38 dogs (48.7%) and 18 dogs had dead puppies. Four records did not state the inhalant.
In 61 dogs (78.2%), conventional uterine incision(s) enabled removal of neonates; 29 dogs had dead puppies. The remaining 13 dogs (16.7%) had an en-bloc removal of the uterus with fetuses; 7 dogs had dead puppies. Five records did not state the technique used. A total of 64 dogs (82.1%) had puppies present in the uterus only; 26 dogs had dead puppies. A total of 14 dogs (17.9%) had a puppy in the pelvic canal and 12 of these dogs had dead puppies.
Discussion
From these results, neonatal mortality was associated with puppies present in the pelvic canal and anesthesia time > 80 min. Other maternal factors were not significantly associated with neonatal mortality, but age > 8 y and time from induction to start of surgery of > 30 min may be possible risk factors.
In this study, risk of neonatal mortality with obstructive dystocia is increased by 8.8-fold. In a previous retrospective study involving 150 records of cesarean section, this risk was also significant and increased neonatal mortality by 2.6-fold (6). Obstructive dystocia with a puppy stuck in the pelvic canal may be due to maternal and fetal factors such as disproportionate size, malpresentation, and uterine inertia (4). These results illustrate the immediate need for emergency surgery when a puppy cannot be removed from the pelvic canal by nonsurgical methods.
A previous study with propofol-isoflurane anesthesia and a 20-minute wait between induction and removal of the puppy, reported the NMR was 21%, and 3% died within 20 min of birth with 26% stillborn (11). Although a wait time of at least 30 min after induction with dexmedetomidine-propofol improved neonate Apgar scores (12), our results indicate periods from induction to start of surgery of < 15 min may favor neonatal survival and periods > 30 min are less favorable. From another study with longer induction to start of surgery time (mean: 35; range: 10 to 70 min), the NMR increased when this period was > 45 min (6). Conversely, in that study, a period < 10 min did not improve neonate survival (5). A limitation of the present study is lack of detail regarding time of removal of the puppy from the uterus. Analysis of anesthesia induction to time of removal of the puppy would enable better evaluation of fetal exposure to anesthetics and NMR.
An anesthesia time > 80 min significantly increased risk by 3-fold and a previous study reported that an anesthesia time of > 2 h (mean: 111 min: range: 45 to 200 min) increased risk by 6.7-fold (6). A retrospective analysis of our results did not show an association with an anesthesia time > 2 h, probably because there were few cases with this length of anesthesia time (P = 0.73) (6). In the previous study, a surgical time > 75 min increased risk 3-fold (6). Although surgical times were similar, surgical time was not a risk factor in our study. Because surgical time was not a risk factor, but anesthesia time was associated with NMR, this could indicate the time from induction to start of surgery is the riskiest period. Long surgery times can be associated with NMR because of multiple uterine incisions for large litters and complications such as uterine rupture. Subsequent spay procedure will also lengthen surgery and anesthesia time but is not associated with NMR, unless the removed puppies become hypothermic and/or hypoglycemic during this time. The difference in surgical time risk between studies may be due to our low sample size because older charts did not contain many surgical details. Retrospective analysis of intraoperative anesthetic complications would also help to further evaluate these findings but was not the focus of this study.
In our study, older dogs had a 7-fold higher risk of having dead puppies. Dystocia was more common in boxer dogs older than 4 y, but NMR was not evaluated (23). Women 40 to 49 y old have higher infant mortality (24). In our study, 2 dogs aged 8 y were reported to have maternal-fetal disproportion, but no details were available for the other dogs. The effect of previous litters was not evaluated in this study and other reasons for age as a possible risk factor cannot be assessed from our results.
Another possible factor linked with increased neonatal mortality was use of sevoflurane facemask induction (P = 0.11). Facemask induction can be used to reduce the dose or eliminate the need for injectable induction drugs (1). However, there is risk of regurgitation/aspiration and increased stress from resistance to the application of a close-fitting facemask, resulting in catecholamine release and decreased uterine perfusion (10,25,26). A shih tzu dog was induced with sevoflurane alone and 2 out of 5 puppies were dead; however, details on the quality of induction were not available. Another well-sedated dog (sufentanil) had an excellent quality of induction with sevoflurane administered via facemask with no anesthetic complications, but 2 puppies died. One other sufentanil-sedated dog induced with sevoflurane had obstructive dystocia as an overlapping risk factor. It must be recognized that the relevance of facemask induction to NMR cannot be confirmed with the low number of dogs involved and overlapping risk factors.
In the present study, opioid administration before induction did not significantly affect NMR, although an increased risk of having one or more dead puppies was suggested by the calculated odds ratio. Although opioids can cause respiratory depression, they are important for maternal perioperative analgesia and reducing the hypotensive effects of inhalants (1,10,27,28). Low concentrations of isoflurane can preserve uterine blood flow; however, levels over twice the minimum alveolar concentration (MAC) can lower uterine perfusion (29). In addition, pregnant animals have a lower MAC compared to nonpregnant animals (30). Hydromorphone and fentanyl with either propofol or alfaxalone appeared to have no effect on neonatal mortality, but it was difficult to separate the risk of using sufentanil from other overlapping risk factors such as facemask induction and obstructive dystocia. Sufentanil has strong binding to alpha 1-acid glycoprotein and neonates have reduced concentrations of this plasma protein. Sufentanil can therefore have prolonged elimination kinetics with a greater proportion of sufentanil available for opioid receptors in neonates (9). Three dogs with dead puppies were sedated with remifentanil and induced with propofol or alfaxalone. Remifentanil has been administered to women for cesarean section without affecting neonatal vigor, but this has not been explored in veterinary medicine (31). Although neonatal resuscitation was not evaluated, inadequate reversal of opioids could increase NMR and affect these results. Naloxone should be administered IV or intranasally, but efficacy of sublingual naloxone has not been evaluated (32,33).
In common with results of a multicenter trial, there was minimal effect on NMR with rapidly eliminated drugs such as propofol and isoflurane (5). The NMR for alfaxalone and propofol induction of anesthesia are similar (14,15) and both can preserve placental perfusion in pregnant ewes (34). Alfaxalone was introduced to Canada in 2011 and results indicate it is a suitable substitute for propofol. Isoflurane and sevoflurane are equally suitable choices to maintain anesthesia during cesarean section and confirms previous findings (17,35).
The NMR of 19.6% from this study compares with the NMR of 20% 1 wk after cesarean section in another study (5). The time interval after cesarean section also appears to increase NMR and the present study included a variable cesarean section to discharge time, which is a confounding factor in calculation of NMR. The time from cesarean section to discharge is affected by dams or puppies requiring intensive care and delays in owner arrival. The time the dams are in labor affects NMR (4) and was not assessed in this study because of factors such as variable recognition of onset of parturition and lack of detail in records.
In our study, the surgical technique employed had no difference on NMR, but the procedures were performed by surgery residents or Boarded surgeons and level of expertise has been shown to improve neonatal vigor (35). In a previous study with 37 dogs, the en-bloc removal technique produced an NMR of 25% with a mean time of arterial ligation to removal of puppy of 40 s (range: 30 to 60 s) (19). In the present study, however, time taken for these steps was not recorded in the medical records for comparison.
Brachycephalic anatomy is also considered to be a risk factor (5). However, most of the brachycephalic dogs in the present study were small and body size was not a factor. Maternal preoxygenation and neonatal oxygen administration alongside other resuscitative efforts remain important but were not assessed in this study (1,3,33). On charts, the preoxygenation box was not always checked, and details of administration and effectiveness were not available. “Flow-by” oxygen administration is not as beneficial as using a close-fitting facemask (36), although the stressful restraint required for the facemask might be counterproductive.
Other limitations inherent to retrospective studies include; data loss, unknown confounding factors, and inability to determine causation only association. During this study period, this clinic transitioned from paper to computerized records and the earlier charts did not have the detail of later charts. The details of previous parturitions, number of dead puppies prior to hospital admittance, length of labor, effect of any anesthetic complications, and accidental death from the dam lying on a puppy can also affect NMR and were not considered.
In conclusion, for dogs undergoing cesarean section, obstructive dystocia and anesthesia time were associated with increased NMR and these results indicate that expediency in decision-making and timely entry into the operating room are important. Large prospective studies are necessary to confirm whether these factors, anesthetic induction regimens, and maternal age can influence NMR.
Acknowledgments
The authors thank the Medical Records staff, Veterinary Medical Centre, Western College of Veterinary Medicine. CVJ
Footnotes
Use of this article is limited to a single copy for personal study. Anyone interested in obtaining reprints should contact the CVMA office (hbroughton@cvma-acmv.org) for additional copies or permission to use this material elsewhere.
References
- 1.Pascoe PJ, Moon PF. Periparturient and neonatal anesthesia. Vet Clin North Am Small Anim Pract. 2001;31:315–341. doi: 10.1016/s0195-5616(01)50208-9. [DOI] [PubMed] [Google Scholar]
- 2.Ryan SD, Wagner AE. Cesarean section in dogs: Physiology and perioperative considerations. Compendium. 2006;28:34–42. [Google Scholar]
- 3.Traas AM. Resuscitation of canine and feline neonates. Theriogenology. 2008;70:343–348. doi: 10.1016/j.theriogenology.2008.04.009. [DOI] [PubMed] [Google Scholar]
- 4.Smith FO. Guide to emergency interception during parturition in the dog and cat. Vet Clin North Am Small Anim Pract. 2012;42:489–499. doi: 10.1016/j.cvsm.2012.02.001. [DOI] [PubMed] [Google Scholar]
- 5.Moon PF, Erb HN, Ludders JW, Gleed RD, Pascoe PJ. Perioperative risk factors for puppies delivered by cesarean section in the United States and Canada. J Am Anim Hosp Assoc. 2000;36:359–368. doi: 10.5326/15473317-36-4-359. [DOI] [PubMed] [Google Scholar]
- 6.Proctor-Brown LA, Cheong SH, Diel de Amorim M. Impact of decision to delivery time of fetal mortality in canine caesarean section in a referral population. Vet Med Sci. 2019;5:336–344. doi: 10.1002/vms3.163. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Potkay S, Bacher JD. Morbidity and mortality in a closed foxhound breeding colony. Lab Anim Sci. 1997;27:78–84. [PubMed] [Google Scholar]
- 8.Kearns GL, Abdel-Rahman SM, Alander SW, Blowey DL, Leeder JS, Kauffman RE. Developmental pharmacology-drug disposition, action, and therapy in infants and children. N Engl J Med. 2003;12:1157–1167. doi: 10.1056/NEJMra035092. [DOI] [PubMed] [Google Scholar]
- 9.Thigpen JC, Odle BL, Harirforoosh S. Opioids: A review of pharmacokinetics and pharmacodynamics in neonates, infants, and children. Euro J Drug Metabol Pharmacokinet. 2019;44:591–609. doi: 10.1007/s13318-019-00552-0. [DOI] [PubMed] [Google Scholar]
- 10.Ryan SD, Wagner AE. Cesarean section in dogs: Anesthetic management. Compendium. 2006;28:44–54. [Google Scholar]
- 11.Funkquist PM, Nyman GC, Löfgren AJ, Fahlbrink EM. Use of propofolisoflurane as an anesthetic regimen for caesarean section in dogs. J Am Vet Med Assoc. 1996;211:313–317. [PubMed] [Google Scholar]
- 12.Groppetti D, Di Cesare F, Pecile A, et al. Maternal and neonatal well-being during elective C-section induced with a combination of propofol and dexmedetomidine: How effective is the placental barrier in dogs? Theriogenology. 2019;129:90–98. doi: 10.1016/j.theriogenology.2019.02.019. [DOI] [PubMed] [Google Scholar]
- 13.De Cramer KGM, Joubert KE, Nöthling JO. Puppy survival and vigor associated with the use of low dose medetomidine premedication, propofol induction and maintenance of anesthesia using sevoflurane gas-inhalation for cesarean section in the bitch. Theriogenology. 2017;19:10–15. doi: 10.1016/j.theriogenology.2017.03.021. [DOI] [PubMed] [Google Scholar]
- 14.Metcalfe S, Hulands-Nave A, Bell M, et al. Multicentre, randomised clinical trial evaluating the efficacy and safety of alfaxalone administered to bitches for induction of anaesthesia prior to caesarean section. Aust Vet J. 2014;92:333–338. doi: 10.1111/avj.12223. [DOI] [PubMed] [Google Scholar]
- 15.Melandri M, Alonge S, Peric T, Bolis B, Veronesi MC. Effects of alfaxalone or propofol on giant-breed dog neonate viability during elective caesarean sections. Animals (Basel) 2019;9:962–973. doi: 10.3390/ani9110962. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Doebeli A, Michel E, Bettschart R, Hartnack S, Reichler IM. Agpar score after induction of anesthesia for canine cesarean section with alfaxalone versus propofol. Theriogenology. 2013;80:850–854. doi: 10.1016/j.theriogenology.2013.07.006. [DOI] [PubMed] [Google Scholar]
- 17.Vilar JM, Batista M, Pérez R, Zagorskaia A, Jouanisson E, Díaz-Bertrana L. Comparison of 3 anesthetic protocols for the elective cesarean-section in the dog: Effects on the bitch and the newborn puppies. Anim Reprod Sci. 2018;190:53–62. doi: 10.1016/j.anireprosci.2018.01.007. [DOI] [PubMed] [Google Scholar]
- 18.Ruiz CC, Del Carro AP, Rosset E, et al. Alfaxalone for total intravenous anaesthesia in bitches undergoing elective caesarean section and its effects on puppies: A randomized clinical trial. Vet Anaesth Analg. 2016;43:281–290. doi: 10.1111/vaa.12298. [DOI] [PubMed] [Google Scholar]
- 19.Robbins MA, Mullen HS. En bloc ovariohysterectomy as a treatment for dystocia in dogs and cats. Vet Surg. 1994;23:48–52. doi: 10.1111/j.1532-950x.1994.tb00442.x. [DOI] [PubMed] [Google Scholar]
- 20.Traas AM. Surgical management of canine and feline dystocia. Theriogenology. 2008;70:337–342. doi: 10.1016/j.theriogenology.2008.04.014. [DOI] [PubMed] [Google Scholar]
- 21.Comazzi S, Pieralisi C, Bertazzolo W. Haematological and biochemical abnormalities in canine blood: Frequency and associations in 1022 samples. J Small Anim Pract. 2004;45:343–349. doi: 10.1111/j.1748-5827.2004.tb00246.x. [DOI] [PubMed] [Google Scholar]
- 22.De Cramer KGM, Joubert KE, Nöthling JO. Hematocrit changes in healthy periparturient bitches that underwent elective ceaseran section. Theriogenology. 2016;86:1333–1340. doi: 10.1016/j.theriogenology.2016.04.075. [DOI] [PubMed] [Google Scholar]
- 23.Forsberg CL, Persson G. A survey of dystocia in the Boxer breed. Acta Vet Scand. 2007;49:8. doi: 10.1186/1751-0147-49-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Donoso E, Carvajal JA, Vera C, Poblete JA. Women age as a risk factor for maternal, fetal, neonatal and infant mortality. Rev Med Chil. 2014;142:168–174. doi: 10.4067/S0034-98872014000200004. [DOI] [PubMed] [Google Scholar]
- 25.Arlt SA. The bitch around parturition. Theriogenology. 2020;150:452–457. doi: 10.1016/j.theriogenology.2020.02.046. [DOI] [PubMed] [Google Scholar]
- 26.Pottie RG, Dart CM, Perkins NR. Speed of induction of anaesthesia in dogs administered halothane, isoflurane, sevoflurane or propofol in a clinical setting. Aust Vet J. 2008;86:26–31. doi: 10.1111/j.1751-0813.2007.00244.x. [DOI] [PubMed] [Google Scholar]
- 27.Mathews KA. Analgesia for the pregnant, lactating and neonatal to pediatric cat and dog. J Vet Emerg Crit Care. 2005;15:273–284. [Google Scholar]
- 28.Hellyer PW, Mama KR, Shafford HL, Wagner AE, Kollias-Baker C. Effects of diazepam and flumazenil on minimum alveolar concentrations for dogs anesthetized with isoflurane or a combination of isoflurane and fentanyl. Am J Vet Res. 2001;62:555–560. doi: 10.2460/ajvr.2001.62.555. [DOI] [PubMed] [Google Scholar]
- 29.Palanuik RJ, Schnider SM. Maternal and fetal cardiovascular and acid-base changes during halothane and isoflurane anesthesia in the pregnant ewe. Anesthesiology. 1974;41:462–472. doi: 10.1097/00000542-197411000-00010. [DOI] [PubMed] [Google Scholar]
- 30.Palanuik RJ, Schnider SM, Eger EI., 2nd Pregnancy decreases the requirement for inhaled anesthetic gases. Anesthesiology. 1974;42:81–83. doi: 10.1097/00000542-197407000-00021. [DOI] [PubMed] [Google Scholar]
- 31.Hill D. Remifentanil in obstetrics. Curr Opin Anaesthesiol. 2008;21:270–274. doi: 10.1097/ACO.0b013e3282f8e26c. [DOI] [PubMed] [Google Scholar]
- 32.Wahler BM, Lerche P, Ricco Pereira CH, et al. Pharmacokinetics and pharmacodynamics of intranasal and intravenous naloxone hydrochloride administration in healthy dogs. Am J Vet Res. 2019;80:696–701. doi: 10.2460/ajvr.80.7.696. [DOI] [PubMed] [Google Scholar]
- 33.Moon PF, Massat BJ, Pascoe PJ. Neonatal critical care. Vet Clin North Am Small Anim Pract. 2001;31:343–367. doi: 10.1016/s0195-5616(01)50209-0. [DOI] [PubMed] [Google Scholar]
- 34.Domínguez E, Rivera del Álamo MM, Novellas R, et al. Doppler evaluation of the effects of propofol, etomidate and alfaxalone on fetoplacental circulation hemodynamics in the pregnant ewe. Placenta. 2013;34:738–744. doi: 10.1016/j.placenta.2013.06.301. [DOI] [PubMed] [Google Scholar]
- 35.Moon-Massat PF, Erb HN. Perioperative factors associated with puppy vigor after delivery by cesarean section. J An Anim Hosp Assoc. 2002;38:90–96. doi: 10.5326/0380090. [DOI] [PubMed] [Google Scholar]
- 36.Ambros B, Carrozzo MV, Jones T. Desaturation times between dogs preoxygenated via face mask or flow-by technique before induction of anesthesia. Vet Anaesth Analg. 2018;45:452–458. doi: 10.1016/j.vaa.2018.03.004. [DOI] [PubMed] [Google Scholar]