Overall this paper is well written, well presented and referenced. The majority of the editorial changes included below are grammatical and intended to help the authors secure publication. However, I have two major comments on the content, which could be rectified via obtaining more field data and adding it to this paper and modifying the statistaical method.

  1. There is a need for additional data to make the conclusions more robust.
  • It is not realistic to base a conclusion of antibiotic resistance on 2 E.coli isolates. There are too few of all other isolates too.
  • Most isolates were A.pyogenes (18) even so an MIC90 would need an absolute minimum of 20 isolates. Any statement of antibacterial resistance even from the simple disk diffusion method (rather than a titration method) needs to be expressed as percent of resistant isolates from a meaningful total number of tested isolates.
  1. Need to clarify how reproductive tract structural abnormalities are graded and use a statistical method to correlate these with bacterial isolation. I suggest the easiest way would be to use a rank system (e.g. 1-4) then use either Pearson’s or Spearman’s Rank testing for strength of correlation.

If the authors could obtain a substantial additional amount of isolates, use a statistical method to assess strength of correlation to support their conclusions and make the suggested grammatical changes this paper could then become suitable for publication.

Bacterial study of clinical postpartum endometritis in Holstein dairy cows

Morteza Yavari1, Masoud Haghkhah2, Mohammad Rahim Ahmadi1

Department of Clinical Sciences1 and Pathobiology2School of Veterinary Medicine, ShirazUniversity, P.O.Box:71345-1731, Shiraz, Iran

aCorresponding Author:

Mohammad Rahim Ahmadi

Professor of Theriogenology

Department of Clinical Sciences, School of Veterinary Medicine, ShirazUniversity,

PO Box: 71345-1731, Shiraz, Iran.

Mobile: +98-917-700-1074

Fax: +98-711-228-6950

E-mail:

Abstract

Endometritis is inflammation of the endometrial lining of the uterus and is associated with delayed uterine involution and poor reproductive performance.The aim of this study was to present the results of bacteriological culture from uterine swabs of dairy cows affected withto with postpartum endometritis and to evaluate the antimicrobial susceptibility of the isolated bacteria. In total, eighty nine Holstein cows affected withto postpartum endometritis were selected and sampled betweenat 21-35 day postpartum. Swabs (n=89) were collected from the uterine lumen of dairy cattle. Bacteria were identified following aerobic and anaerobic culture and the disk diffusion method was performedused to determine susceptibility of the major pathogenic isolated bacteria. The results revealed that the most common isolates from cases of endometritis in studystudied cows were Arcanobacterium pyogenes, E. coli, and non-differentiated streptococci, staphylococciand bacilli (type?)us. The antimicrobial susceptibility tests showed that E. coli were sensitive to enrofloxacin and ceftiofur (only 2 isolated means nothing), but they were resistant to tetracycline and oxytetracycline. For A. pyogenes, 72, 66, 72 and 72 percent of isolates were resistant to oxytetracycline, tetracycline, enrofloxacin and cCeftiofur respectively. All isolates showed resistance to penicillin. In conclusion, using of oxytetracycline which is as the most traditional route of antimicrobial therapy for postpartum endometritis in cows appears not to be efficacious. There is widespread resistance to It may be true for enrofloxacin and third generation of cephalosporins tooas well. Therefore, it seems the dairy farms would need to study and evaluate alternatives to treating and preventing post-partum endometritis includingthe non antibiotic options.alternative drugs for the treatment of postpartum endometritis.

Keywords: Endometritis; Bacteria; Antimicrobial susceptibility; Uterus; Dairy cow

Abbreviations:

EPC=Epithelial cells

LVEP=large vacuolated epithelial cells

MAC=macrophage

EN=endometritis

Neut=neutrophils

Lym=lymphocytes

LH= Lutenizing hormone

C=Cervix

Introduction

During parturition, the physical barriers of the cervix, vagina and vulva are compromised providing the opportunity for bacteria to ascend the genital tract. Bacteria can be isolated from the uterus of over 90% of cows early postpartum (Griffin et al., 1974; Paisley et al., 1986). Most healthy cows are able to clear the uterus of bacteria within the first 2 to 3 weeks after calving (Bondurant, 1999). However, cows that cannot eliminate the infection may subsequently develop endometritis (Dhaliwal et al., 2001).

Endometritis is a common reproductive disorder in female domestic animals with consequences ranging from no effect on reproductive performance to permanent sterility. It affects the general health of animals and adversely affects their reproductive performance (Amiridis et al., 2003). Subclinical endometritis, based on uterine cytological examination, is also prevalent in dairy cows and has a profound negative impact on reproductive performance (Hammon et al., 2006). The presence of bacteria in the uterus causes inflammation, histological lesions of the endometrium and delays uterine involution (Bonnett et al., 1991,Sheldon et al., 2003). In addition, uterine bacterial infection or bacterial products suppress pituitary LH secretion, and perturb postpartum ovarian follicle growth and function, which disrupts ovulation in cattle (Sheldon et al., 2002b, Opsomer et al., 2000,Peter and Bosu., 1988,Peter et al., 1989). Thus, endometritis is associated with lower conception rates, increased intervals from calving to first service or conception, and more culls for failure to conceive (Borsberry and Dobson, 1989,Huszenicza et al., 1999, LeBlanc et al., 2002).

In the world, a A variety of antimicrobial agents, administered by intrauterine infusion or parenteral injection, are regularly used to treat uterine infections (Cohen et al., 1995). There is little reportliterature available concerningabout bacterial causes of postpartum endometritis and their susceptibility to suitable candidate antibiotics in Iranian dairy farms. The aim of this study was: (a) to identify isolates bypresent bacteriological culture from uterine swabs of Iranian Holstein dairy cows with clinical endometritis between 21 and 35 days postpartum that affected by clinical postpartum endometritis and (b) to evaluate the antimicrobial susceptibility of the most common isolates.the principal isolated bacteria from cases of endometritis in the dairy cows.

MATERIALS AND METHODS

Animals

The study was carried out in 13 largebig commercial dairy herds of Iran (please give range of cow numbers). Four hundred two postpartum dairy cows were examined once between 21 and 35 days postpartum. In total 89 cows affected with clinical endometritis affected cows were selected. Cows in all herds were calved in calving boxes hygienically and kept in individual boxes for at least 10 days after parturition. Corn silage, alfalfa hay, and concentrates as a total mixed ration were used.None of the cows received any intrauterine or reproductive hormonal therapy for at least 10 days before sampling for this study.

Clinical examination

During the examination, Ccows were first inspected examined for the presence of recentfresh discharge onon the vulva, perineum, or tail. If discharge was not visible externally, cows were examined vaginally.The cow’s vulva was thoroughly cleaned with a dry paper towel and a clean, lubricated, gloved hand was inserted throughinto the vaginavulva. In each cow, the lateral, dorsal and ventral walls of the vagina were palpated, and the mucus contents of the vagina withdrawn manually for examination, as described by (Sheldon et al..,(2002a). The vaginal mucus was assessed for color and proportionpresence of pus. The nature of the discharge was classified as clear mucus, clear mucus with flecks of pus, mucopurulent (approximately 50% pus and 50% mucus), purulent (>50% pus) but not foul-smelling, purulent or red-brown and foul smelling using the methodology described by(LeBlanc et al. (, 2002).

Following vaginal examinationinspection, transrectal palpation of the reproductive tract was performed and cervical diameter, location of the uterus, symmetry of the uterine horns, diameter of the (larger) uterine horn, texture of uterine wall, palpable uterine lumen, dominant palpable ovarian structure including corpus luteum (CL), follicle, cyst (>2.5 cm in diameter), or no palpable structures was recorded (LeBlanc et al., 2002). (do you mean palpable abnormalities? – no structure means absence of tract palpated)

Ultrasonographic assessment of uterus and ovaries using a 5MHz rectal linear probe (AMI Company, Canada) was also performed. Diameter of the uterus, echotexture and thickness of the uterine wall and intraluminal fluid accumulation were evaluated in the cows. Ovarian structures (follicle, CL and cyst) were scanned and measured by calipers (Mateus et al., 2002).

Uterine swab collection and bacteriological culture

For each animal, a transcervical guarded swab was collected from the uterine body (Noakes et al., 1989). The swab comprised a long copper wire bearing a cotton wool tip sheathed in a metal guard tube (8 mm external diameter; 58 cm long) and was wrapped and sterilized by autoclaving at 121°C for 15 min. The guard tube was covered by a sterile plastic sheath to prevent contamination of the swab during the cervix insertion.

After restraining the animal and securing its tail, the perineal region was washed and cleaned. The cervix was grasped per-rectum and the sterilized catheter was passed through the cervix into the uterine body. Then, the inner rod of the catheter was pushed forward to expose the swab to the endometrium and was rotated against the uterine wall and then withdrawn within the catheter. To avoid contamination, the catheter was then cleaned with alcohol. Swabs were cultured immediately on sheep blood agar andMacConkey agar (MERCK), and incubated at 37°C for 48 h. The same culture on sheep blood agar (MERCK) was incubated anaerobically for up to 7 days. Standard biochemical tests were used for the isolation and identification of the isolates as described by (Quinn et al. (, 1994).

Blood sampling and progesteroneP4 assays

Blood samples were collected from the coccygeal vein or artery into evacuated tubes and transported on ice to the laboratory. Serum was separated by centrifugation at 2500 rpmx g for 10 min and stored frozen at ­20°C until required. Plasma progesterone concentration was measured by radioimmunoassay (Spectria® Progesterone RIA, Espoo, Finland) with a sensitivity of 0.1 ng/ml and intra- and inter-assay coefficients of variation of 10.2% and 6.5 % respectively.

Antimicrobial susceptibility tests

According to the categorization of bacteria isolated by culture of uterine swabs, based on their potential pathogenicity within the uterus (Sheldon, 2004), Arcanobacterium pyogenes and E. coli were the only major pathogens identified and used for antibiotic susceptibility tests. Disk diffusion method was performed to determine susceptibility of the major isolated pathogenic bacteria based on the NCCLS 1996 protocol (Please cite the reference as NCCLS 1996 etc and add to list at end of paper). The bacterial suspension turbidity adjusted to McFarland standard number 0.5, in Mueller-Hinton broth (MERCK) and culturedinoculated fluently over the entire surface of on Mueller-Hinton agar (MERCK) with a sterile cotton swab.

Commercial antibiotic disks containing single concentrations of each antibiotic were then placed onto the inoculated plate surface. Inhibition zone of growth around each disk after overnight incubation at 37°C, were measured. The zone diameter was interpreted using a zone size interpretation chart (Lorian, 1996). The antibiotics and their concentration per disc were as follows: tetracycline 30µg, oxytetracycline 30µg, penicilin 10 international units (iu)Units, enrofloxacin 5µg and ceftiofur 30µg (Quinn et al., 1994).

Statistical analysis

Data were analyzed by using SAS software, version 6.12. Association of ovarian structure (please detail how), progesteroneP4level of serum level and discharge status (is this ranked by severity) with bacterial culture results in postpartum endometric cows were determined by using Chi-square and Fisher exact tests (SAS, 1991).

Results

In total, 89eighty nineHolstein cows were selected and sampled at 21-35 day postpartum. Thirty five (39.3%) swabs were found bacteriologically positive and the remaining 54 (60.7%) showed no bacterial growth. Aerobic and facultative anaerobic bacteria which were isolated are listed in Table 1. A total of 61 isolates were identified from the positive swabs. The most frequently isolated facultative anaerobe was A. pyogenes 18 (29.51%) followed by Bacillus spp. 13 (21.31%), Streptococci8 (13.11%), Staphylococci (9.83) and Lactobacillus 8 (13.11%).Twenty three (66%) of positive swabs yielded pure bacterial growth, of which A. pyogenes was the most frequently isolate (Table 1). Character of uterine discharge and bacterial isolation variables were dependent and mucus characters had significant different bacterial isolation (P<0.05). Character of uterine discharge and bacterial isolation variables were dependent and mucus characters had significant (P<0.05) different bacterial isolation (Table 2). Ovarian structure and P4 level of serum were not correlated with independent from bacterial isolation (P>0.05) (Table 3).(Should use Spearmans or Pearson’s rank correlation coefficients)

The result of antimicrobial susceptibility tests showed that both of E. coli isolates (not enough to be conclusive) were sensitive to enrofloxacin and ceftiofur, but they were resistant to tetracycline and oxytetracycline (Table 4). For A. pyogenes, (state number of isolates as not many and % is misleadingly conclusive) 72, 66, 72 and 72 percent of isolates were resistant to oxytetracycline, tetracycline, enrofloxacin and ceftiofur respectively. All bacteria isolate showed resistanceto penicillin (Fig.1).

TABLE 1. Bacterial isolates from postpartum endometritis of dairy cows

Bacteria / Number / %
A. pyogenes / 18 / 29.51
E. coli / 2 / 3.28
Streptococci (total) / 8 / 13.11
S. equinus / 1 / 1.64
S. dysgalactiae / 1 / 1.64
S. spp / 6 / 9.83
Staphylococci (total) / 6 / 9.83
S. epidermidis / 1 / 1.64
S. spp / 5 / 8.19
Corynebacterium spp / 2 / 3.28
Mannhiema haemolytica / 1 / 1.64
Micrococcus spp / 3 / 4.92
Bacilli (total) / 13 / 21.31
B. coagulans / 3 / 4.92
B. firmus / 3 / 4.92
B. pumilus / 3 / 4.92
B. spp. / 4 / 6.56
Lactobacillus spp / 8 / 13.11
Total / 61 / 100

Table 2. VThe compare of vaginal mucus discharge status and bacterial isolation in postpartum endometritis cows.

Character of mucus
Bacteriology / Clear mucus / Mucus with flecks of pus / Mucopurulent / Purulent / total
Positive / 9(10.11)a / 5(5.62)abc / 10(11.24)b / 11(12.36)c / 35(39.33)
Negative / 34(38.2)a / 11(12.36)abc / 7(7.87)b / 2(2.25)c / 54(60.68)
total / 43(48.31) / 16(17.98) / 17(19.10) / 13(14.61) / 89(100)

Values with different superscripts in each row are those that differ significantly (p<0.05)

Table 3. Association of ovarian structure and plasma progesterone concentrationP4level of serum with with results of bacterial culture in postpartum endometritis cows

Bacteriology / Ovarian Structure
Nos. (%) / P4
Nos. (%)
No structures / Follicle / Corpus luteum / < 1ng/ml / > 1ng/ml
Positive / 4 / 22 / 9 / 26 / 9
Negative / 1 / 35 / 18 / 34 / 20
total / 5 (5.62) / 57 (64.04) / 27 (30.34) / 60 (67.4) / 29 (32.6)

(what does negative structure mean??)

Table 4. The antimicrobial susceptibility tests of the major isolated pathogenic bacteria from endometritis cows by using disk diffusion method.

E. Coli
(number of isolates = 2!) / Arcanobacterium pyogenes
(number of isolates = 18) / Bacteria & Results
Sensitive% / Intermediate% / Resistance% / Sensitive% / Intermediate% / Resistance% / Antibiotics
0 / 0 / 100 / 11 / 17 / 72 / oxytetracycline
0 / 100 / 0 / 11 / 23 / 66 / tetracycline
0 / 0 / 100 / 0 / 0 / 100 / penicillin
100 / 0 / 0 / 28 / 0 / 72 / ceftiofor
100 / 0 / 0 / 28 / 0 / 72 / enrofloxacin

You do not need both Fig 1 and Table 4 as they show the same data use one or other – if correlation coefficients calculated preference is probably for the table.

Fig. 1: Antibiotic susceptibility of isolated A. pyogenes from uterine swabs

Discussion

The uterine lumen is sterile before parturition and if bacterial invasion occurs, there is usually resorption of the fetus or abortion (Semambo et al., 1991). During or shortly after parturition, microorganisms from the animal's environment, skin, and feces contaminate the uterine lumen (Sheldon, 2004; Sheldon and Dobson, 2004; Földi et al., 2006). The flora of the postpartum uterus has been shown to be quite variable, and the results of one sample may not give a full picture of the infection status (Griffin et al., 1974). However, the intrauterine bacterial infection as such, does not even necessarily present asmean a clinical manifestation of disease; this is dependant on the immune status of the host (Földi et al., 2006).

The results of this study revealed that the most common isolates from cases of endometritis in cattlewere A. pyogenes, E. coli (no only 2 is not one of the most common!), streptococci (8) and staphylococci.(6) The species of isolated bacteria were similar to those reported in the previous studies (Elliot et al., 1968; Sagartz and Hardenbrook, 1971; Griffin et al., 1974; Hartigan et al., 1974; Studer and Morrow, 1978; Ball et al., 1984; Messier et al., 1984). – this last sentence cannot be used as a conclusion there is insufficient data. In our study, the most common isolate was A. pyogenes. Series of studies confirmed that most of the clinical and reproductive consequences are attributed to the presence of certain non-specific pathogens: mainly to A. pyogenes, either alone or in combination with other bacteria such as E. coli and gram negative obligate anaerobes (Földi et al., 2006). Isolation of A. pyogenes at the late involution period (28-35 days after calving) is associated with dramatically decreased re-conception rate (Huszenicza et al., 1999). In addition; A. pyogenes was the most frequently single pure isolate (40%) in our study.Previous studies have shown A. pyogenesto be recovered in pure culture more frequently with increased days postpartum (Hartigan et al., 1974).

Obligate anaerobes such as Fusobacterium necrophorum and Prevotella spp were not isolated in this study. This result is in agreement with some of other published reports. Kaczmarowski et al. (2004) reported that anaerobes were of lower significance in their examinations. In our opinion, it seems that the inconsistency of the bacterial isolates in different studies may be related to the different bacterial flora of each area and country.However, Williams et al. (2005) explained that variation between studies may include the use of clinically ill animals or the selection of animals for microbiological examination based on likely uterine disease. In addition, the inclusion of animals <21 days postpartum in other studies may be important as they have a different bacterial profile to those calved longer.

The effect of character of vaginal mucus on bacterial isolation was considered independently in the present study. Positive results from bacterial culture was more commonly associated with mucopurulent or purulent vaginal mucus (give correlation co-efficient if you wish to make this conclusion). These observations are in agreement with the previous result of Williams et al. (2005). They reported that the bacterial growth densities for recognized and potential pathogens were associated with purulent vaginal mucus, and higher growth density for recognized pathogens was associated with mucopurulent mucus character. Pus is formed as a result of bacterial infection by a mixture of viable and dead neutrophil leucocytes, necrotic tissue and tissue fluid, so it is not surprising that bacterial growth density of pathogenic bacteria is associated with purulent vaginal mucus (Williams et al., 2005).