I imagine most people will agree that perinatal and neonatal medicine in companion animal practice lacks a certain degree of sophistication and expertise when compared to human medicine. The reasons for this are of course numerous and not for elaboration here. I did recently come across someone referring to the modified APGAR score when describing a case of dystocia in a bitch and I must admit I had not heard of it being used before.
So what did I do? Well obviously I went to see what papers I could find to try and help me answer questions such as:
- What is the modified APGAR score? Is there one version or several?
- How has the human APGAR score been modified to achieve this?
- To which species has it been applied?
- What is the evidence base? Has it been tested and validated? If so, how? Prospectively in a RCT?
- And of course a big question which is what implications if any does it have for clinical practice? Have I been missing something useful all these years?! In my experience the vast majority of puppies (just focusing on puppies here) that die in the perinatal period do so during parturition or very quickly after birth. What will a modified APGAR score help me do differently?
- Would it for example allow us to give puppies with subjectively questionable viability some time to monitor progress after prompt medical intervention rather than immediately performing euthanasia?
- Would it prompt us to be more aggressive with our care of neonates with questionable viability? But what level of care can we realistically provide, or be allowed to provide, for these cases? How many dog carers will invest in intensive medical care for neonates?
Remembering of course that the purposes of scoring systems – suitably validated – are for example to:
- Help objectify assessment and facilitate monitoring (especially when different assessors are involved) – this could be of an individual organ dysfunction o the entire patient
- Help prognosticate/predict outcome
- Help guide therapy and resource allocation
- Facilitate clinical trials
- Help to evaluate clinical performance
In human medicine the APGAR score may correlate with short-term mortality but it is not meant to be a predictor of long-term neurological outcome in survivors. It is intended to guide doctors in providing care to vulnerable newborn babies. But it should also be remembered that other parameters such as umbilical vein lactate and cardiotocography may be used as well of course as clinical observation.
I wouldn’t claim to have done a long and exhaustive search by any means but below are the papers I came across that applied to companion animals; there are also published papers relating to horses and farm animals. And for anyone that has not come across the ‘Appearance Pulse Grimace Activity Respiration’ score before (where have you been?!) let me link you to the all-knowing Wikipedia
for a quick summary of this system which is used to assess the health of babies immediately after birth. A brief human medicine literature search seems to suggest that this scoring system is the most widely used system and continues to be used; there is at least one modification as well, the Surgical APGAR score.
Veronesi MC, Panzani S, Faustini M, Rota A. An Apgar scoring system for routine assessment of newborn puppy viability and short-term survival prognosis. Theriogenology 2009. 72:401-407.
193 puppies from 42 litters
Different breeds with a good distribution of small, medium and large
Birth by spontaneous whelping, assisted whelping or caesarean section; good even spread across the three groups
Survival classified as:
- Born dead
- Born alive but died within 2 h
- Born alive, still alive at 2 h and rechecked at 24 h
Modified APGAR score – see Figure 1. Performed within 5 mins of birth.
Reflex irritability was tested by gentle compression of the tip of a paw and evaluating the degree of reaction with respect to vocalisation and leg retraction
- 27/193 puppies born dead, leaving 166 alive
- 3/166 were immediately euthanised leaving 163 puppies alive
- 157 puppies were alive at 24 h; 6 died despite 20 minutes of resuscitation attempts
- Of the 6 that died, 4 were in APGAR score group 0-3 – but note that 3 other puppies in this group survived
- Of the 6 that died, 2 were in APGAR score group 4-6 – but note that 15 other puppies in this group survived
- The vast majority (85.3%) of puppies were in the APGAR score group 7-10
- Medical treatment was provided to puppies with APGAR score less than 6
- Puppies with lower APGAR scores also, unsurprisingly, showed less mammary gland searching and suckling and swallowing reflexes
The authors write: “Statistical analysis showed that the percentage of puppies that were dead at 2 h was higher in the 4 to 6 Apgar score group compared with that in the 7 to 10 score group (P < 0.01). This difference was even greater when the 0 to 3 group was compared with the 7 to 10 score group (P < 0.0001).”
Without taking the lid of that can of worms here (not least as I am not that good at statistics to dare to expand!), the ‘P value’ has come under quite some critique in terms of what it means and how it has come to be used in medicine. All is not well in these regards so don’t assume a statistically significant P value necessarily means something useful and especially not something clinically useful!
I guess my take on this paper was that its prospective nature was good but the lack of a control group was a shame, i.e. a study where decisions to provide treatment were in some cases determined by APGAR score and in other cases standard clinical assessments. The lack of a control group is raised by the authors. They attribute recovery of some of the puppies with APGAR score < 6 to the medical treatment given and this may well have been the case but clinicians who do not use a modified APGAR scoring system will decide which puppies need medical treatment and to what extent based on similar assessments just not in a ‘score’. Without a control group, there is no mention of how big the sample size needed to be to be powered to show a real effect of APGAR scoring; this lack of an adequate sample size to power a study is widespread in veterinary studies even where control groups are present. While 193 puppies might sound like a decent number, the fact that so few were in the lower APGAR score groups means that the utility here seems less and bigger group numbers would be needed in my view.
The authors acknowledge that they did not follow up survival beyond 24 h which is obviously the more interesting thing to know as puppies can die for up to 2 weeks (or more obviously). However they say that after the immediate perinatal period other factors beyond initial viability come into play and this is not the intended use of the APGAR score in humans either, i.e. it is really just about early neonatal viability. This seems fair to me.
So after reading this paper, I looked at some others and will mention them in a lot less detail.
Batista M, Moreno C, Vilar J, et al. Neonatal viability evaluation by Apgar score in puppies delivered by cesarean section in two brachycephalic breeds (English and French bulldog). Anim Repro Sci 2014. 146:218-226.
“The aims of the present study were to define the neonatal survival and birth defects in puppies delivered by cesarean section in brachycephalic breeds (English Bull Dog, French Bull Dog). In addition, the reliability of an adapted Apgar score as a tool to check newborn viability and to predict neonatal survival was also assessed.”
302 puppies from 44 litters of 40 bitches
They used the modified APGAR score in the paper above to evaluate neonatal viability: of course this demonstrates nicely one of the potential uses of a scoring system but we need to bear in mind how well the system has been validated before employing it in this way.
They used a different classification for final APGAR score: 0-3 = critical neonates; 4-6 = moderate viability; 7-10 = normal viability.
APGAR score was assessed within the first 5 mins after delivery and about 60 mins later
APGAR scoring was always done by the same 2 people and the final score was derived from the average of their assessments.
Overall more of the puppies proportionately were in the lower APGAR score groups than in the study above.
Spontaneous neonatal mortality by 24 h after birth was < 5% and the vast majority of these were in the APGAR score 0-3 group; survival for puppies with APGAR score ≥ 4 was close to 100%.
They state “Our results who that there is a close relationship between the Apgar score and neonatal viability prognosis and while there is no guarantee that all those pups showing a good Apgar score will automatically survive, it seems clear that the puppies with high Apgar scores are more likely to survive.” Fair enough!
Doebeli A, Michel E, Bettschart R et al. Apgar score after induction of anesthesia for canine caesarean section with alfaxalone versus propofol. Theriogenology 2013. 80:850-854.
“The objectives of this study were to evaluate, in a clinical setting, the effects of alfaxalone as an anesthetic induction agent for dogs undergoing emergency C-section and to compare neonatal viability after alfaxalone or propofol anesthetic induction.”
They had 11 bitches in each group, i.e. alfaxalone vs. propofol; this led to 81 puppies of which 73 were alive at birth.
They used the modified APGAR score in the paper above to evaluate neonatal viability.
APGAR score was assessed at 5, 15 and 60 minutes after delivery.
Results relating to use of APGAR score of relevance to this blog (i.e. I am not delving into the anaesthetic drug aspect here) were that APGAR score increased in both groups from 5 to 60 minutes after delivery.
Across both groups, this study had a greater proportion than the one above in the lower APGAR score groups.
They felt that “use of the Apgar scoring system adapted for puppies by Veronesi et al. at 5, 15, and 60 minutes after delivery allowed for objective comparison between anesthetic drugs.”
Groppetti D, Pecile A, Del Carro AP, et al. Evaluation of newborn canine viability by means of umbilical vein lactate measurement, apgar score and uterine tocodynamometry. Theriogenology 2010. 74:1187-1196.
So as the title states these investigators attempted to evaluate neonatal viability in a more comprehensive way beyond just APGAR score, similar to some human medicine practice.
They used a different modified APGAR scoring system to the one reported above with 7 categories, each being allocated a score of 0 to 2: mucous membrane colour; heart rate; respiratory rate; reflex of irritability; mobility; suckling; and, vocalisation.
- Total score: 0-14
- 0-4 = severely stressed; 5-9 = moderately stressed; 10-14 = healthy
- Done within 10 mins of delivery and repeated 24 h and 48 h after birth – so they included one later assessment point which is no bad thing!
It can be seen that compared to the one above this system separates out ‘respiratory effort’ into its 2 components as separate categories, i.e. respiratory rate and vocalisation, and also includes a suckling category that the one above did not.
See Figure 2
The APGAR system here was adapted for puppies from one proposed in another paper for horses in 2007. As far as I can tell, this was de novo and this system had not been tested elsewhere – however I think I like it better than the Veronesi one.
Only 21% of the newborn pups were classified as healthy using their system.
This paper obviously was looking at other things beyond APGAR score not described in this blog but as far as the APGAR score the main point here is “No statistical differences were detected between alive pups and those dead within 48 h in relation to APGAR score assigned at birth.” “APGAR scores at delivery were not related to pup survival within 48 h of birth”. This therefore contradicts what the papers above reported. We need to note that the papers above stopped their assessments at 24 h and we do not know what happened after that but there is no suggestion in this paper here that APGAR score correlated to 24 h survival even if not to 48 h survival.
I did also come across another paper but did not feel that it contributed anything noteworthy to this review and especially not when critically appraised (Silva LCG, Lucio CF, Veiga GAL, et al. Neonatal clinical evaluation, blood gas and radiographic assessment after normal birth, vaginal dystocia or caesarean section in dogs. Reprod Dom Anim 2009. 44 (Suppl. 2):160-163.).
So what did I take from all this jibber jabber?
I guess my take is that I would not use a modified APGAR score to decide whether or not to provide medical treatment to a puppy – but I would fully anticipate that most of the puppies to which I felt the need to provide medical treatment would be in the lower APGAR score groups. But some in the high/healthy group may still need care.
I would also not use this score to definitively prognosticate on a puppy’s likely survival outcome – although clearly the ones in the lowest/worst group are intuitively less likely to survive than the ones in the highest group.
I don’t believe one can make definitive clinical decisions based on either of the scoring systems reported above, more and better studies evaluating the use of the modified APGAR score are needed before that becomes a rational path to go down in my view.
However I certainly can see how a modified APGAR score would be useful as a supplementary monitoring tool to inform clinical observations and assess clinical progress. And no doubt the next time I am face to face with a live neonatal puppy I will assess his/her modified APGAR score – and probably using the one reported in the Groppetti paper.
If anyone does using a modified APGAR score I would love to hear from you, and also of course if you have any comments on the above!