Accepted Manuscript

How to perform transabdominal ultrasound-guided fetal fluid sampling in mares

Igor F. Canisso, DVM, MSc, DACT, DECAR Barry A. Ball, DVM, PhD, DACT EdwardL. Squires, MS, PhD, DACT (hon) Mats H. Troedsson, DVM, PhD, DACT, DECAR

PII: S0737-0806(14)00254-8

DOI: 10.1016/j.jevs.2014.06.013

Reference: YJEVS 1746

To appear in: Journal of Equine Veterinary Science

Received Date: 1 April 2014

Revised Date: 28 May 2014

Accepted Date: 13 June 2014

Please cite this article as: Canisso IF, Ball BA, Squires EL, Troedsson MH, How to performtransabdominal ultrasound-guided fetal fluid sampling in mares, Journal of Equine Veterinary Science(2014), doi: 10.1016/j.jevs.2014.06.013.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service toour customers we are providing this early version of the manuscript. The manuscript will undergocopyediting, typesetting, and review of the resulting proof before it is published in its final form. Pleasenote that during the production process errors may be discovered which could affect the content, and alllegal disclaimers that apply to the journal pertain.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

1

How to perform transabdominal ultrasound-guided fetal fluid sampling in mares 1

Igor F. Canisso DVM, MSc, DACT, DECAR; Barry A. Ball DVM, PhD, DACT*; Edward L. 2

Squires MS, PhD DACT (hon); Mats H. Troedsson DVM, PhD DACT, DECAR; 3

Reproduction Laboratory, The Maxwell H. Gluck Equine Research Center, Department of 4

Veterinary Science, University of Kentucky, Lexington KY 40546, USA. 5

6

*Corresponding author at: Barry A. Ball DVM, PhD, DACT, Maxwell H. Gluck Equine 7

Research Center, Department of Veterinary Science, University of Kentucky, 8

108 Gluck Equine Research Center, Lexington, KY 40546-0099, USA. 9

Email address: b.a.ball@uky.edu 10

11

Abstract 12

Little is known about the composition and physiology of fetal fluids in all domestic 13

mammals in comparison to humans, where the amniotic fluid has been the focus of numerous 14

reports. Previously, in the horse there have been concerns regarding the safety of fetal fluid 15

sampling and the risks to the well-being of the fetus and techniques limitations that may preclude 16

serial assessment. The objective of this report is to describe a transabdominal ultrasound-guided 17

technique to safely perform multiple fetal-fluid collections during the last trimester of gestation 18

in mares. Similar methodology has been described previously; however, here we described step 19

by step how to perform fetal fluid sampling. Several illustrative images have been included to 20

facilitate the understanding of the technique by others lacking experiences with the procedures. 21

In addition, small modifications on the sedation protocols and sampling have been performed in 22

the present study. Six light horse mares carrying normal, singleton pregnancies (260-280 days of 23

gestation) were sampled three times at five to seven day intervals (i.e. 0, 5 and 12d). There were 24

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

2

no apparent complications using the protocol described here. All mares delivered normal foals 25

uneventfully. We foresee that the publication of this report may be useful to other research 26

laboratories interested in studying the fetal fluids in mares, as well as in specific clinical 27

evaluations of fetal well-being in research mares. 28

Key-words: allantois, amnion, pregnancy, ultrasound, equine, centesis 29

30

1. Introduction 31

32

In domestic mammals, particularly in the horse, the physiology of fetal fluids is poorly 33

understood. Few studies have been conducted to examine the physiopathology and chemical 34

composition of the equine fetal fluids [1-8], in comparison to humans, where the amniotic fluid 35

has been the focus of numerous reports [9-10]. Initial attempts to collect equine fetal fluids were 36

carried out before the wide use of ultrasonography, and resulted in numerous adverse outcomes 37

(i.e. abortion and even peritonitis). Later attempts using ultrasonography to guide the collections 38

reduced the risks of abortion [4-5], but some of the protocols using transabdominal 39

ultrasonography to locate pockets of equine fetal fluids are still associated with abortion [1]. As 40

an alternative method, Lyle and others [6] reported the placement of an indwelling catheter in the 41

allantoic compartment by laparoscopy in attempt to allow continuous fetal fluid sampling. 42

Unfortunately, several complications were reported (e.g. lack of fixation of the catheter, infection 43

around the catheter and chorioallantois, and abortion) and the catheters remained patent for only 44

5-9 days due to blockage by cell debris. 45

Transabdominal ultrasound-guided fetal fluid sampling appears to be the safest method 46

[5]. Recently, our group has performed multiple transabdominal ultrasound-guided fetal fluid 47

sampling, as part of ongoing research to examine proteomics of the fetal fluids in normal and 48

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

3

placentitis mares [11], though we have made some modifications in a successful previously 49

described protocol [5] to obtain fetal fluid samples during the third trimester of pregnancy. Our 50

uneventful and successful outcomes stimulated us to describe here in great detail a protocol for 51

transabdominal ultrasound guided fetal fluid sampling, which is currently in use at our laboratory 52

to collect repetitive samples of equine fetal fluids. 53

2. Material and Methods 54

The experimental protocol (# 2010-0769) was approved by the Institutional Animal Use 55

and Care committee at the University of Kentucky. This project was carried out from April to 56

August of 2012, at the Maine Chance Farm, University of Kentucky, Lexington, Kentucky, 57

USA. Six light horse mares carrying normal, singleton pregnancies (260-280 days of gestation) 58

were sampled three times at five to seven day intervals (i.e. 0, 5 and 12d). To assure that the 59

procedures did not induce any damage in the placentas, and as part of another study, all placentas 60

were submitted for microbiologic and pathologic examination at the University of Kentucky 61

Veterinary Diagnostic Laboratory. None of the mares received any medication throughout 62

gestation, other than sedation and local anesthesia (see below) for the fetal fluid sampling. 63

Before each fetal fluid sampling, the mare was given xylazine hydrochloride 64

(0.4mg/kg/iv; Anased®, Lloyd Laboratory, Shenandoah, Iowa, USA) and allowed to rest quietly 65

(~5min) to facilitate handling, clipping, and ultrasonography. Once sedated, the mare was moved 66

slowly to the stock to avoid stress and excitement. 67

Ethanol (70%) was spread through the mare’s ventral abdomen to facilitate the 68

visualization of the pregnant uterus via transabdominal ultrasonography (curvilinear transducer 69

2–6 MHz, C362 Sonoscape Sonosite Bethel, Washington, USA). If needed the mare was clipped 70

before ultrasonography. Ideally a large area should be clipped prior to ultrasonography. For the 71

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

4

centesis, a combination of sedation and analgesia was provided by additional administration of 72

detomidine hydrochloride (0.008 mg/kg/iv; Dormosedan®, Pfizer Animal Health, Exton, 73

Pennsylvania, USA) and butorphanol tartrate (0.008 mg/kg/iv; Turbogesic®, Pfizer Animal 74

Health, Fort Dodge, Iowa, USA). 75

A large area of the mare’s ventral abdomen was surgically prepared with chlorhexidine 76

scrub and ethanol 70%. To decide where to perform local anesthesia, a pocket of fetal fluid was 77

located by ultrasound. Once the puncture site was located, lidocaine hydrochloride (i.e. 10-78

15mL; Lidocaine hydrochloride®, VEDCO, Saint Joseph, Missouri, USA) was injected (23-25G 79

2’’) in the subcutaneous tissues and body wall (Fig 1). The transducer was placed inside a sterile 80

sleeve containing sterile lube. Two operators were needed, and strict aseptic technique was used 81

to perform fetal fluid sampling. 82

The operator sampling the mare held the transducer in one hand, and the echotip spinal 83

needle (18G x 6”; 30o short bevel, Chiba-Type spinal needle, coupled with a stylet, Echo Block® 84

PTC Havel’s Cincinnati, Ohio, USA) in the other hand (Fig 2A). At this point the operator had 85

confirmed the location of a pocket of fetal fluid. Thereafter, the needle was inserted through the 86

body wall and uterus into the fetal fluid compartments (Figs 2A, 3C, 3D). The amniotic 87

compartment could be located as an echogenic membrane surrounding the fetus; by 88

transabdominal ultrasound the operator could localize the pocket of amniotic fluid (anechoic 89

appearance) between the fetal parts and the amniotic membrane. Whereas the allantoic fluid is 90

the fluid present between the uterine wall and amniotic membrane, it could be more easily 91

accessed caudally closer to the mare’s bladder. Midline or paramedian approaches did not 92

appear to be essential, but rather how a pocket of fetal fluid was accessible to the operator. After 93

needle placement, the second operator on the opposite side of the mare removed the stylet and 94

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

5

connected the extension set with the syringe to the needle (Fig 2B). Very gentle suction was 95

applied by the operator to obtain fetal fluids (Fig 2C). Once the amount of fetal fluid desired was 96

collected the needle was rapidly removed (Fig 2D). Afterwards the mare was allowed to recover 97

from sedation in a quiet stall or paddock without hay or water. Fig 3 (A-D) illustrates all the 98

different steps from sedation to sampling. 99

100

3. Results 101

There were no apparent complications using the protocol described here. All mares 102

delivered normal foals uneventfully. Four mares passed their placentas ≤3h post-partum. One 103

mare after retaining her placenta (~6h), was treated with intravenous bolus of oxytocin (100 104

units, diluted in 1L of Lactate Ringer’s Solution), delivered her placenta (~45min) after 105

treatment [12]. Another mare retained her placenta for 4d and was treated with daily large uterine 106

flushing, oxytocin, cloprostenol, sulfamethoxazole and trimethoprim, and flunixin meglumine 107

[12]. Post-partum pathologic placental examinations were unremarkable, except one placenta 108

that had advanced autolysis as a consequence of the prolonged retention. 109

4. Discussion 110

Here we described a minimally invasive procedure for fetal fluid sampling that has been 111

successfully used in our laboratory. Although this study had limited number of mares, there 112

were no apparent short or long term complications. Despite 2 mares had retained their placentas 113

(one for short period of time and other for prolonged period of time), we cannot be certain 114

whether the retentions were related to fetal fluid sampling, however, it is worth to note that both 115

mares have had history of having retained fetal membranes in previous uneventful parturitions. 116

Mares having retained placentas without a predisposing factor (e.g. dystocia) tend to repeat in 117

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

6

future pregnancies [12]. Though, if retained placentas were induced by fetal fluid sampling, this 118

can be considered minimal compared to more invasive procedures (e.g. laparoscopic placement 119

of an indwelling catheter) [6] or abortion [1]. In addition, in a parallel study serum amyloid A, a 120

sensitive marker of inflammation, which is elevated upon placental infection [13], was not 121

elevated in association with fetal fluid sampling [14]. As aforementioned, throughout gestation 122

and up to the second stage of parturition, none of the mares received any medication other than 123

sedation and local anesthesia required for the fetal fluid sampling, and all mares carried normal 124

pregnancies and had uneventful parturition, except two mares had placental retention as 125

discussed above. These findings indicate that the procedures here described did not induce 126

detectable inflammation or cause major clinical problems that could jeopardize pregnancy. 127

Some of the techniques described in the literature to sample equine fetal fluids require 128

skin incisions [3] which could be a limiting factor in studies aimed to address inflammation, for 129

example in mares with experimentally induced placental pathology. In contrast, the protocol 130

described here can be used to safely collect equine fetal fluids multiple times if properly applied 131

without inducing systemic inflammation. As noted above, our protocol is similar to the 132

methodology previously described by Paccamonti and colleagues [5]; however, there were some 133

differences in the sedation protocol and in the type of needle used as discussed below. 134

In our hands the use of spinal needle with an echotip appears to facilitate the sampling, as 135

the operator could be certain where the needle was placed and more accurately sampling of the 136

desired fetal fluid compartment. Spinal needle coupled with a stylet has been previously reported 137

[4,5], however this is the first report to use the needle with echotip and also to attach an 138

extension set to a syringe, this probable allows faster sampling and clean sampling of the desired 139

volume. Previously, other authors [3-5] collected fetal fluids by gravity, where the spinal needle 140

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

7

had to be repositioned multiple times when spontaneous drainage of the fetal fluid was 141

discontinued, this probable explain why some authors reported multiple holes in the amniotic 142

membrane [4]. 143

Earlier studies, the authors either used detomidine and butorphanol [4] or detomidine 144

alone [5] or xylazine and butorphanol [3]. Herein we used a combination of these three drugs, 145

while we cannot be certain whether the previously reported sedation protocols would have 146

different outcome than ours; it was beyond the scope of this report to compare sedation 147

protocols. Alternatively to the sedation protocol that we applied in the present report, the 148

detomidine and butorphanol administration can be given to the mare before the she was placed in 149

stocks, especially if noted that the mare did not respond to the xylazine. Previous experiences 150

working with the mare may guide the clinician to decide, whether she needs sedation with 151

detomidine and butorphanol before she has been hand-walked into the stocks. Commonly, 152

before all the procedures are completed such type of mare will need further sedation with 153

detomidine and butorphanol, which we usually give the half or a third of the whole dose, 154

dependent upon the mare’s previous behavior and how she reacts to sedation. 155

Worth noting, at this stage of pregnancy equine fetal fluids are very distinct from one 156

another; amniotic fluid is a transparent liquid with minimal visible flocculent material, whereas 157

allantoic fluid is dark-yellow-tan (i.e. urine like) appearance (Fig 2D). It was outside the scope of 158

this report to present data on the chemical composition of samples, but others have used to 159

confirm the origin of the equine fetal fluids [4,5], Allantoic fluid is rich in creatinine, presents 160

higher content of sodium and chloride and has a low content of calcium compared to the 161

amniotic fluid. Cytological appearance of these two fetal fluids is also very distinct, as the 162

amniotic fluid (Fig 4A) presents higher cellularity in comparison to the allantoic fluid (Fig 4B). 163

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

8

Research groups interested in performing fetal fluid sampling in mares, ideally should 164

have available a transabdominal-ultrasound machine coupled with a curvilinear transducer 2–6 165

MHz that has effective image depth of ≥20-25cm; that allows the operator to appreciate different 166

reference points that cannot be easily observed in ultrasound machines with less effective image 167

depth . For example, if the amniotic fluid is the desired fluid compartment, the operator can 168

localize the fetal shoulder; between the fetal neck and head the operator will often see a pocket 169

of amniotic fluid that can be relatively easily assessed (Fig 3C); however, ultrasound machines 170

with low effective image depth (usually small portable machines), will not easily allow full 171

appreciation of such reduced pockets of fetal fluids. On the other hand if the operator is aiming 172

for the allantoic fluid, the operator can use the cervical star area and the mare’s bladder to 173

position the transducer to obtain a good view of a pocket of fetal fluid inside this compartment 174

(Fig 3D). 175

During fetal fluid samplings, we observed that there was a midway point in the mare’s 176

ventral abdomen that makes fetal fluid compartments more difficult to reach, as it is the lowest 177

point of the abdomen, thus it appeared to us that the uterus was more difficult to reach with the 178

spinal needle, however, if the operator moves the transducer slightly cranial or caudal such 179

difficulty can be easily overcome. 180

In summary, the procedures described herein appear to be safe to perform multiple fetal 181

fluid samplings guided by ultrasound. The combination of a sedation protocol, associated with a 182

transabdominal ultrasound machine with an effective depth ~20cm, and the use of spine needle 183

with an echotip appears to facilitate the procedures. We foresee that the publication of this report 184

may be useful to other research laboratories interested to study equine fetal fluid, as well as in 185

specific clinical evaluations of fetal well-being in research mares. 186

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

9

Acknowledgements 187

188

This study was supported by the Kentucky Thoroughbred Association and Kentucky 189

Thoroughbred Breeders and Owners and by the University of Kentucky (Department of 190

Veterinary Science, Geoffrey Hughes Fellowship, Albert Clay Endowment). We would like 191

express our gratitude to Dr Jan Grove of the “Foal In Mare Team” at the University of Ghent, 192

Belgium, for preparing the drawings on Fig 3 (A-D). None of the authors have any conflict of 193

interest or relation with any third part that bias the publication of this report. Sidney Hughes is 194

thanked for her assistance during the execution of this study. 195

196

References 197

[1] Schmidt AR, Williams MA, Carleton CL, Darien BJ, Derken FJ. Evaluation of 198

transabdominal ultrasound-guided amniocentesis in the late gestation. Equine Vet J 1991; 23: 199

261-265. 200

[2] Smith KC, McGladdery AJ, Binns MM, Mumford JA. Use of transabdominal ultrasound-201

guided amniocentesis for detection of equid herpesvirus 1-induced fetal infection in uteri. Am J 202

Vet Res 1997; 58:997- 1002. 203

[3] Williams MA, Wallace SS, Tyler JW, McCall CA, Gutierrez A, Spano JS. Biochemical 204

characteristics of amniotic and allantoic fluid in late gestational mares. Theriogenology 1993; 205

40:1251-1257. 206

[4] Holdstock NB, McGladdery AJ, Ousey J, Rossdale PD. Assessing methods of collection and 207

changes of selected biochemical constituents in amniotic and allantoic fluid throughout equine 208

pregnancy. Biol Reprod Monograph 1995; 1: 21-38. 209

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

10

[5] Paccamonti DL, Swiderski CE, Marx B, Gaunt S, Blouin D Electrolytes and biochemical 210

enzymes in amniotic and allantoic fluid of the equine fetus during late gestation. Biol Reprod 211

Monograph Series 1995; 1: 39-48. 212

[6] Lyle SK, Paccamonti DL, Huberta JD, Schlafer DH, Causey RC, Eilts B, Johnson JR. 213

Laparoscopic placement of an indwelling allantoic catheter in the mare: biochemical, cytologic, 214

histologic, and microbiologic findings. Anim Reprod Sci 2006; 94:428-431. 215

[7] Zanella LF. Biochemical analyses of the amniotic and allantoic fluids throughout gestation in 216

Equus caballus. Dissertation thesis, São Paulo State University. Botucatu, São Paulo, Brazil 217

2008. 218

[8] Zanella LF, Takahira RK, Melo-Ona CM, Ona-Magalhaes LC, Prestes NC. Biochemical 219

profile of amniotic and allantoic fluid during different gestational phases in mares. J Equine Vet 220

Sci 2014; 34:403-406. 221

[9] Scott F, Peters H, Boogert T, Robertson R, Anderson J, McLennan A, Kesby G, Edelman D. 222

= The loss rates for invasive prenatal testing in a specialized obstetric ultrasound practice. Aust 223

N Z J Obst Gynaecol 2002; 42: 55-58. 224

[10] Mujezinovic F, Alfirevic Z. Procedure-related complications of amniocentesis and chorionic 225

villous sampling: a systematic review. Obst Gynecol 2007; 110:687-694. 226

[11]Canisso IF, Ball BA, Scoggin KE et al. Alpha-fetoprotein is highly expressed and appears to 227

be increased in the foetal fluids of mares with placentitis. In Proceedings. European Society for 228

Domestic Animal Reproduction. Reprod Domestic Anim 2013; 48(supplement S1): 116. 229

[12] Canisso IF, Rodriguez JS, Sanz MG, Coutinho da Silva MA. A clinical approach to the 230

diagnosis and treatment of retained fetal membranes with an emphasis placed on the critically ill 231

mare. J Equine Vet Sci 2013; 33:570-579. 232

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

11

[13] Coutinho da Silva MA, Canisso IF, Macpherson ML, Johnson AE, Divers TJ. Serum 233

amyloid A concentration in healthy periparturient mares and mares with ascending placentitis. 234

Equine Vet J 2013; 45:619-624. 235

[14] Canisso IF, Ball BA, Cray C, Scoggin KE, Davolli GM, Squires EL, Troedsson MH. Serum 236

Amyloid A and Haptoglobin Concentrations Are Elevated in Plasma of Mares with Ascending 237

Placentitis in the Absence of Changes in Peripheral Leukocyte Counts or Fibrinogen 238

Concentration. Am J Reprod Immunol (accepted). 239

240

Fig.1: Local lidocaine block guided by ultrasound. Note, the transducer was held where the 241

operator found a readily accessible pocket of fetal fluid. 242

Fig. 2: (A) Note that the operator was holding the transducer covered with a sterile sleeve in one 243

hand, while the spine needle was being inserted with the other hand. This approach allowed the 244

operator to visualize the pocket of fetal fluid; therefore a very accurate puncture could be carried 245

out. (B) Operator #1 was holding the needle on the mare’s right side, while operator #2 located 246

on the mare’s left side had connected a syringe with an extension set to the needle. (C) Operator 247

#1 was holding the needle on the mare’s right side, while the operator #2 aspirated the allantoic 248

fluid. (D) Allantoic fluid immediately after sampling. Note the needle with its stylet. The stylet 249

was removed from the needle lumen once the operator #1 considered that the needle was placed 250

in the desired spot. Upon the stylet removal, often but not always, the fetal fluid drained 251

spontaneously through the needle. 252

253

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

12

Fig. 3: (A) representation of a late term pregnant mare before sedation; (B) Note that once the 254

mare has been sedated her head and neck are lowered, as well as a consequence of sedation the 255

abdominal muscles are relaxed which brought the fetus closer to the abdominal ventral body, this 256

fetal positioning associated with relative quiescence mare’s status facilitated the sampling. (C) 257

Representation of the procedure during amniocentesis, note the transducer helps to keep track of 258

the echo-tip needle transposing the body wall. (D) Drawing representative of an allantocentesis, 259

similarly to C, the transducer was kept aligned to the needle to help the operator’s to coordinate 260

the needle. Images were kindly prepared by Dr. Jan Govaere of the “Foal In Mare” project at the 261

University of Ghent, Belgium. 262

Fig 4: Cytologic slides the fetal fluids prepared by cytospin and stained by Diff-Quick. (A) 263

amniotic fluid was very rich in epithelial type cells and presented relative scant debris among 264

cells; (B) allantoic fluid was acellular, remarkable amorphous material was observed. 265

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT