Contribution of Unmeasured Anions to Acid–Base Disorders and its Association with Altered Demeanor in 264 Calves with Neonatal Diarrhea

Abstract

Background

The quantitative effect of strong electrolytes, unmeasured anions (UAs), p CO ₂, and plasma protein concentrations in determining plasma pH and bicarbonate (HCO ₃ ⁻) can be demonstrated using the physicochemical approach. Demeanor of calves with diarrhea is associated with acidemia, dehydration, and hyper‐d‐lactatemia.

Hypothesis

Unmeasured anions are a major factor influencing changes in plasma pH and HCO ₃ ⁻ of calves with diarrhea and UAs and strong UAs, estimated by anion gap (AG) and strong ion gap (SIG), respectively, are more strongly associated with alteration of demeanor compared to other acid–base variables.

Animals

A total of 264 calves with diarrhea from two data sets (DS1 and DS2).

Methods

Retrospective study. Forward stepwise regression was performed to determine the relationship between measured pH or HCO ₃ ⁻, and physicochemical variables. A two‐way ANOVA was performed to investigate the association between acid–base variables and attitude (bright, obtunded, and stuporous), posture (standing, sternal or lateral recumbency), and strength of suckling reflex (strong, weak, or absent).

Results

Increased strong UAs estimated by SIG was the most important contributor to changes in measured pH and HCO ₃ ⁻ (DS1: r ² 66 and 59%, DS2: 39 and 42%, P < .0001). SIG and AG were correlated to deteriorating calf demeanor for all three clinical scoring categories: attitude, posture, and suckle reflex (P < .0001).

Conclusion and Clinical Relevance

Elevated concentrations of strong UAs were the primary cause of acidemia and had an important influence on the demeanor of calves with diarrhea. These findings emphasize the importance of the calculation of UAs when evaluating acid–base abnormalities in calves.

Abbreviations

A⁻

total net negative charge of blood proteins

AG

anion gap

A_tot

total plasma concentration of nonvolatile weak acids

BE_(ecf)

base excess extracellular fluid

d‐lac⁻

d‐lactate

HCO₃⁻

bicarbonate

H–H

Henderson–Hasselbalch approach

l‐lac⁻

l‐lactate

mmol

millimoles

pCO₂

partial carbon dioxide pressure

SID

strong ion difference

SIG

strong ion gap

TP

total protein

UAs

unmeasured anions

Neonatal diarrhea is a major cause of illness and death in calves less than 1 month of age.1 Diarrhea can lead to death from dehydration, acidemia, hyperkalemia, and impaired cardiovascular and renal function, regardless of the causative pathogens or pathophysiologic mechanisms.2, 3, 4 Historically the most commonly stated causes of metabolic acidosis in diarrheic calves were fecal bicarbonate loss, electrolyte and water loss, and l‐lactic acidosis.5, 6 More recently, acidemia in sick calves, with or without diarrhea, was primarily attributed to hyponatremia accompanied by normochloremia or hyperchloremia, increased strong unmeasured anions particularly d‐lactate ([d‐lac⁻]), and to a lesser extent increased plasma protein concentrations.7

Two approaches have been used to evaluate acid‐base imbalances, the Henderson–Hasselbalch approach (H–H)8 and more recently the physicochemical approach.9, 10 The H–H centers on plasma bicarbonate ([HCO₃ ⁻]) and extracellular base excess (BE_(ecf)) concentrations as indicators of the metabolic components of acid–base balance. This approach is clinically useful for assessing acid–base disorders; but it is more descriptive than mechanistic and ignores the importance of relative strong electrolyte and protein concentrations in the determination of acid–base balance.11, 12 The quantitative physicochemical approach emphasizes the importance of strong electrolytes (Na⁺, K⁺, Cl⁻), strong unmeasured anions (including l‐lactate⁻ [l‐lac⁻] and d‐lactate [d‐lac⁻]), pCO₂, and the total plasma protein concentration (A_tot) in determining plasma pH and [HCO₃ ⁻].9 This approach is mechanistic and provides additional information regarding the pathophysiology of acid–base imbalances.

Altered demeanor is associated with the degree of hypoglycemia, metabolic acidosis, hypothermia and dehydration. Hypoglycemia is a concurrent disease in calves with acute severe diarrhea.13, 14 Hypoglycemia is associated with alteration of demeanor such as impaired concentration, focal neurological deficits, confusion, drowsiness, coma, seizure, and neuronal death in different species.13, 14

Assessment of the degree of metabolic acidosis on the basis of clinical signs has been investigated by several authors.3, 5, 15, 16, 17, 18 A calf's demeanor is influenced by the severity of metabolic acidosis, hypothermia and degree of dehydration. Furthermore, dehydration correlates with blood [l‐lac⁻] concentration.19 There is an association between BE_(ecf) and [HCO₃ ⁻] and changes in the posture, behavior, and suckling reflex of calves with diarrhea.15, 20 Increased serum concentration of [d‐lac⁻] occurs in calves with diarrhea with or without signs of dehydration.21, 22, 23 The variations in behavior and posture of diarrheic calves can be better explained by an increase in serum [d‐lac⁻] rather than by a decrease in BE_(ecf).24 In support of this, alterations in behavior and posture, impairment of the palpebral reflex, somnolence, and staggering gait occurs after intravenous administration of sodium [d‐lac⁻] or isotonic d‐l‐lactate to healthy calves25; whereas experimentally induced severe, uncomplicated, hyperchloremic metabolic acidosis does not result in changes in posture or mental status.26 These studies indicate that metabolic acidosis without an appreciable increase in [d‐lac⁻] had only a minor influence on posture and behavior.18 The routine measurement of plasma [d‐lac⁻] is not currently a common practice in veterinary medicine. Clinicians assume that other frequently calculated acid–base parameters such as anion gap (AG) are sufficient for predicting changes in the concentration of the unmeasured anions. A useful clinical application of the physicochemical approach is calculation of the strong ion gap (SIG). SIG represents the difference between unmeasured strong ion cation concentration and unmeasured strong ion anion concentration in plasma or serum. SIG treats total plasma protein and phosphate concentration as variables, whereas AG assumes they are constants.11 Both AG and SIG were intended to identify the presence of increased concentrations of unmeasured anions such as d‐ and l‐lactate and uremic acids.27

We hypothesized that increased plasma concentration of unmeasured anions are the major independent variable influencing changes in measured plasma pH and calculated plasma HCO₃ ⁻ and that UAs and strong UAs, estimated by AG and SIG, respectively, are more strongly associated with alteration of demeanor in calves with diarrhea compared to other acid–base variables. The objectives of this study were (1) to determine and characterize the mechanism of acid–base disorders in calves with diarrhea using the physiochemical approach and (2) to investigate the correlation between the degree of metabolic acidosis, using parameters from the H–H and physicochemical approaches, and alteration in calf demeanor in a population of calves with naturally occurring diarrhea.

Materials and Methods

Study Population

The analysis was based on 2 data sets (DS1 and DS2). DS1 was derived from 171 calves ≤28 days of age that were admitted to the Atlantic Veterinary College Teaching Hospital (AVC‐TH) with a primary complaint of diarrhea between 1989 and 1993, and that had venous blood gas and biochemical profile values determined upon admission.28 DS2 consisted of data extracted from the medical records of 83 calves ≤28 days of age, admitted to the AVC‐TH between January 2000 and December 2010, with a diagnosis of neonatal diarrhea, and that had blood gas analyses and biochemistry profiles performed upon admission.

Data Collection

Medical records were systematically reviewed and the following signalment information was recorded: breed, sex, and age in weeks (week 1: calves ≤7 days old, week 2: calves 8–14 days old, etc.). Recorded physical examination findings were as follows: attitude (bright, obtunded, or stuporous), posture (standing, sternal or lateral recumbence), suckle reflex (strong, weak, or absent), heart rate (HR) as beats/minute (bpm), respiratory rate (RR) as breaths/minute (rpm), injection of the sclera, rectal temperature (T) in degrees Celsius (C), and state of dehydration estimated based on clinical signs by the attended clinician (5–6%, mild dehydration; 7–9%, moderate dehydration; >10%, severe dehydration). The following clinicopathologic data were extracted: p _vCO₂ (mmHg), pH, [HCO₃ ⁻] (mmol/L), BE_(ecf) (mmol/L), [Na⁺] (mmol/L), [K⁺] (mmol/L), [Cl⁻] (mmol/L), [TP] (g/L), and creatinine concentration (μmol/L). An antemortem diagnosis of septicemia in a diarrheic calf was based on any one or more the following criteria28: (1) positive blood culture; (2) culture of the same bacterial agent from ≥2 body fluids; and (3) culture of bacterial agent from a single joint in a calf with joint effusion involving multiple joints. A postmortem diagnosis of septicemia was based on the presence of one or more of the following criteria: (1) morphologic changes such as multiple disseminated abscesses of similar size, purulent vasculitis and intravascular identification of bacteria, or fibrin in multiple body cavities; (2) bacterial isolation from heart blood; or (3) recovery of the same bacterial organism from ≥2 body tissues (excluding intestine). The pathogens causing diarrhea were also recorded. Duration of hospitalization was based on either the number of days until the calf was discharged from the hospital or on the number of days until death. Survival was defined as discharge from the hospital.

Measurement Techniques

A radiometer ABL‐11 blood gas machine with autocalibrating systems and a portable IRMA true point blood gas analyzer2 were used for blood gas and pH analysis for DS1 and DS2, respectively. Serum biochemical profile and electrolytes analyses were performed using a Coulter Dacos analyzer3 and a Copas 6000 C501 automated multianalyzer4 system for DS1 and DS2, respectively. Total protein concentration was measured using refractometry.

Calculations

The H–H's equation was used to calculate [HCO₃ ⁻] (mmol/L) and BE_(ecf) (mmol/L) from the measured values for venous pH and pCO₂, and assigned values for S of 0.0306 mmol/L/mmHg and pK₁ of 6.099 mmol/L and for S of 0.0307 mmol/L/mmHg and pK₁′ of 6.095 for DS1 and DS2, respectively. Therefore, [HCO₃ ⁻] and BE_(ecf) were calculated as:

$$[{\text{HCO}}_3^{-}]=S\times p{\text{CO}}_2\times {10}^{({\mathrm{pH}}_{\mathrm{m}}-p{{\mathrm{K}}_1}^{\text{'}})}$$ (1)

$${\text{BE}}_{(\mathrm{ecf}\mathrm{)}}={\text{[HCO}}_3^{-}]-24.8+16.2\times ({\text{pH}}_{\mathrm{m}}-7.40)$$ (2)

BE_(ecf) assumes a fixed hemoglobin concentration of 50 g/L.4

Determination of the strong ion difference (SID) requires identification and accurate measurement of all strong ions in plasma, including ideally the measurement of unmeasured strong anions. SID has two components: measured SID, which was calculated from the measured plasma concentrations of 3 strong ions (Na⁺, K⁺, Cl⁻) as9:

$${\text{SID}}_3=\text{(Na+K‐Cl)}$$ (3)

and unmeasured SID (SID_um) calculated as:

$${\text{SID}}_{\mathrm{um}}=S\times p{\text{CO}}_2\times {10}^{(\mathrm{pH}-p{{\mathrm{K}}_1}^{\text{'}})}-{\mathrm{A}}_{\mathrm{tot}}/[1+{10}^{(p{\mathrm{K}}_{\mathrm{a}}-\mathrm{pH}\mathrm{)}}]-{\text{SID}}_3$$ (4)

using assigned values for S of 0.0307 mmol/L/mmHg and pK₁′ of 6.120.7

A useful clinical application of the simplified SID approach is the calculation of strong ion gap (SIG), which provides an estimate for the difference between the ion charge of plasma nonvolatile buffers and unmeasured strong anion concentration.7 Rearrangement of the Equation (4) and SIG can be substituted for SID_um as7:

$$\text{SIG}=[{\mathrm{A}}_{\mathrm{tot}}]/[1+{10}^{(p{\mathrm{K}}_{\mathrm{a}}-\mathrm{pH}\mathrm{)}}]-\text{AG}$$ (5)

Anion gap (AG) was calculated as29:

$$\text{AG}(\text{mmol}/\mathrm{L})=([{\text{Na}}^{+}]+[{\mathrm{K}}^{+}])-([{\text{Cl}}^{-}]+[{\text{HCO}}_3^{-}])$$ (6)

Total plasma concentration of weak acids (A_tot) was calculated as:

$${\mathrm{A}}_{\mathrm{tot}}(\text{mmol}/\mathrm{L})=[0.343\times (\text{TP})]$$ (7)

where total protein is in grams per liter.7

Total negative charge of the plasma proteins A⁻ (mmol/L) were calculated as:

$${\mathrm{A}}^{-}(\text{mmol}/\mathrm{L})=[{\mathrm{A}}_{\mathrm{tot}}]/(1+{10}^{(p{\mathrm{K}}_{\mathrm{a}}-\text{pH})})$$ (8)

where pK_a (7.08) is the effective dissociation constant of bovine plasma weak acids.7

Definitions

The definition of acid–base disorders by using the H–H and physicochemical approaches was presented in the Table 1.7, 30

Table 1.

Definition of acid–base disorders using Henderson–Hasselbalch (H–H) and physicochemical (SID) approach.

Approach	Type of Disorder	Parameter7, 27	Acidosis	Alkalosis
H–H	Respiratory	p vCO2 [34–45 mmHg]	↑pvCO2	↓p vCO2
	Metabolic	HCO3 − [20–30 mmol/L]	↓HCO3 −	↑HCO3 −
		AG <20 mmol/L	↑AG	N/A
SID	Respiratory	p vCO2 [34–45 mmHg]	↑p vCO2	↓p vCO2
	Metabolic	SID3 [38–44 mmol/L]	↓SID3	↑SID3
		Atot [19–25 mmol/L]	↑Atot	↓Atot
		SIG [3 to −3 mmol/L]	↓SIG	N/A

Modified from Constable.11 p _vCO₂, partial carbon dioxide pressure; HCO₃ ⁻, bicarbonate; AG, anion gap; SID₃, strong ion difference measured; SIG, strong ion gap; A_tot, total plasma concentration of nonvolatile weak acids.

This article is being made freely available through PubMed Central as part of the COVID-19 public health emergency response. It can be used for unrestricted research re-use and analysis in any form or by any means with acknowledgement of the original source, for the duration of the public health emergency.

Statistical Analyses

Data were presented as a mean and SD. Non‐normal distributed data were presented as a median and quartiles 25 and 75%. Normality of the data was tested by the Kolmogorov–Smirnov test. Forward stepwise regression (P < .05 to enter and exit) was performed to determine the relationship between measured plasma pH and SID₃, SIG, pCO₂, and A_tot for DS1 and DS2, and to determine the relationship between calculated plasma HCO₃ ⁻ and SID₃, SIG, pCO₂, and A_tot for DS1 and DS2. The assumptions of the forward stepwise regression procedure were evaluated by examining the residual plots. The results of the regression were confirmed using backward elimination. To evaluate the association between the acid–base variables and alteration of calf demeanor, both DS1 and DS2 were compiled as a single data set. Before combining them, the 2 data sets were analyzed separately and residual standard deviation of the variables of interest was compared and was found to be similar. A two‐way ANOVA was performed to investigate the association between acid–base variables from the H–H (pCO₂, BE_(ecf), [HCO₃ ⁻] and AG) and the physicochemical approach (pCO₂, SID₃, SIG and A_tot), and attitude, posture, and strength of the suckle reflex. The additive effect of DS1 and DS2 also was investigated in the ANOVA test but it did not have statistical significance (P > .05). Bonferroni correction for pairwise comparison was performed when indicated. The appropriateness of the results of the ANOVA test was evaluated by the examination of residual plots. A P‐value <.05 was considered significant. Statistical analyses were performed using statistical software.5

Results

Study Population

Of the 264 calves, 51% were female, and 49% were male. Breed distribution was similar to that for the general hospital population, which included both beef and dairy calves in both data sets. The median age of calves was 8 days (range 1–28 days). Cases were presented during all 12 months of the year, with the highest numbers seen between January and April. A final etiologic diagnosis of the diarrhea was achieved in 137 (52%) calves. Enterotoxigenic Esherichia coli was the causative agent of diarrhea in 9% (n = 11) of calves less than 5 days of age. Coronavirus was detected in 29% (n = 38) of calves and a total of 26% (n = 34) of calves tested positive for Cryptosporidium spp. One calf tested positive for Salmonella spp, whereas 2% (n = 7) were positive for Eimeria spp. Forty‐six (33%) calves with diarrhea tested positive for more than one pathogen (Rotavirus and Coronavirus n = 15, Rotavirus and Cryptosporidium spp. n = 9, and Coronavirus and Cryptosporidium spp. n = 22).

Physical Examination Findings, Acid–Base, and Electrolyte Abnormalities and Outcome

All calves had diarrhea at the time of hospitalization. Median rectal temperature was 38.5°C (Q₂₅–Q₇₅, 37.1–39.3°C), median HR was 120 bpm (Q₂₅–Q₇₅,100–136 bpm), median RR was 32 rpm (Q₂₅–Q₇₅, 24–48 rpm). Eight percent (n = 20) of calves were considered to have normal hydration status, 26% (n = 69) calves had 5–6% dehydration, 33% (n = 87) had 7–9% of dehydration, and 33% (n = 88) calves had dehydration >10%. Evaluation of the posture of the calves revealed that 33% (n = 88) were standing, 36% (n = 92) were in sternal recumbence, and 31% (n = 84) were in lateral recumbence. Attitude assessment showed that 17% (n = 44) of the calves were bright, 64% (n = 169) were obtunded, and 19% (n = 51) were stuporous. Evaluation of the suckle reflex showed that 23% (n = 62) of calves had a strong suckle reflex, 42% (n = 110) had a weak suckle reflex, and 35% (n = 92) had an absent suckle reflex. The values for the venous blood gas analysis, plasma concentration of electrolytes and plasma proteins, and H–H and physicochemical variables of the 264 calves with diarrhea are presented in Tables 2 and 3. On admission 12% (n = 31) calves had clinical evidence of pneumonia, 7% (n = 18) of arthritis, 1% (n = 3) of meningitis, and 21% (n = 55) of omphalophlebitis, and 33% (n = 86) of the calves were considered to be septicemic based on the previously described criteria. Seventy percent of the calves (n = 184) were discharged, whereas 30% (n = 80) died or were euthanized because of poor prognosis or economic constraints. Out of 80 calves that did not survive, 69% (n = 55) were septic and 31% (n = 25) were nonseptic.

Table 2.

Admission plasma concentration of electrolytes, total plasma protein, albumin and creatinine of 264 calves with neonatal diarrhea.

Parameter (reference range)	Low (%)	Normal (%)	High (%)
Sodium [132–152 mmol/L]31	39 (15)	206 (78)	19 (7)
Potassium [3.9–5.8 mmol/L]31	24 (9)	178 (68)	62 (23)
Chloride [95–110 mmol/L]31	45 (17)	180 (68)	39 (15)
Total protein [57–81 g/dL]31	36 (14)	205 (77)	23 (9)
Albumin [21–35 g/dL]31	7 (3)	251(95)	6 (2)
Creatinine [67–195 μmol/L]31	N/A	174 (61)	90 (39)

This article is being made freely available through PubMed Central as part of the COVID-19 public health emergency response. It can be used for unrestricted research re-use and analysis in any form or by any means with acknowledgement of the original source, for the duration of the public health emergency.

Table 3.

Admission values of 264 calves with neonatal diarrhea with and without sepsis obtained on admission.

Parameter	Calves (n = 264)	Reference Range7, 27
H–H model
pH	7.15 (7.05–7.29)	7.35–7.50
pCO2 (mmHg)	50.3 (40.42–58.8)	34–45
HCO3 − (mmol/L)	18 (11.3–25.8)	20–30
BE(ecf) (mmol/L)	−10.2 (−18.9 to 0.3)	−2.5 to 2.5
AG (mmol/L)	22.5 (16.8–27.9)	14–20
SID model
SID3 (mmol/L)	41.2 (36.8–46.5)	38–44
SIG (mmol/L)	−10.8 (−16.8 to −3.11)	−3 to 3
Atot (mmol/L)	21.9 (19.2–24.6)	13–25

Data presented as median (25th and 75th quartiles). H–H, Henderson–Hasselbalch approach; SID, physicochemical approach; HCO₃ ⁻, bicarbonate; pCO₂, partial carbon dioxide pressure; BE_(ecf), base excess extracellular fluid; AG, anion gap; SID₃, strong ion difference; SIG, strong ion gap; A_tot, total plasma concentration of nonvolatile weak acids.

This article is being made freely available through PubMed Central as part of the COVID-19 public health emergency response. It can be used for unrestricted research re-use and analysis in any form or by any means with acknowledgement of the original source, for the duration of the public health emergency.

Determination of Acid–Base Disorders in Calves with Diarrhea Using Henderson–Hasselbalch and Physicochemical Approaches

Analysis of acid–base balance using the H–H approach (HCO₃ ⁻) detected acid–base alterations in 95% (n = 251) of the calves included in this study; acidosis was present in all (n = 251) of these calves. Eighteen percent (n = 46) and 82% (n = 205) of the calves, had a single or mixed acid–base disorder, respectively (Fig 1).

Figure 1.

Interpretation of acid–base disorders on admission for each calf using Henderson‐Hasselbalch approach. AG, anion gap; p CO ₂, partial carbon dioxide pressure.

Physicochemical analysis of acid–base balance revealed acid–base disorders in 98% (n = 259) of the calves included in this study; acidemia was present in 100% (259) of these calves. Strong ion gap acidosis was the most common disorder and was detected in 80% (n = 206) of the calves included in this study. The interpretation of the acid–base disorders on admission of each calf by the physicochemical approach is presented in Figure 2.

Figure 2.

Interpretation of acid–base disorders on admission for each calf using the physicochemical approach. SIG strong ion gap; p CO ₂ partial carbon dioxide pressure; SID, strong ion difference; A_tot, total concentration of plasma proteins.

Determination of the Mechanism of Acid–Base Abnormalities in Calves with Diarrhea Using the Physicochemical Approach

The results of the forward stepwise regressions with venous pH and calculated HCO₃ ⁻ as the dependent variables and venous SID₃, SIG, pCO₂, and A_tot values as independent variables are presented in Tables 4 and 5. The regression analysis revealed that an increased unmeasured strong anions concentration (SIG) was the most important contributor to the changes in the measured pH and calculated plasma HCO₃ ⁻. A decrease in SID₃ concentration was the next most significant contributor to the changes in the measured pH and calculated plasma HCO₃ ⁻. Changes in A_tot and pCO₂ had a minor but significant contribution to the changes in measured pH and calculated plasma HCO₃ ⁻.

Table 4.

Results of the forward stepwise regression of measured pH as dependent variable versus jugular venous values of the physicochemical variables.

Order of entry into regression model dS1	Variable	Partial R 2	Model R 2
1	SIG	0.66	0.66
2	SID3	0.11	0.77
3	pCO2	0.09	0.86
4	Atot	0.05	0.91
Order of entry into regression model DS2
1	SIG	0.39	0.39
2	SID3	0.25	0.64
3	pCO2	0.17	0.81
4	Atot	0.12	0.93

DS1, data set 1; DS2, data set 2; pCO₂, partial carbon dioxide pressure; SID₃, strong ion difference; SIG, strong ion gap; A_tot, total plasma concentration of nonvolatile weak acids. P < .0001 for all the variable's coeffficients.

This article is being made freely available through PubMed Central as part of the COVID-19 public health emergency response. It can be used for unrestricted research re-use and analysis in any form or by any means with acknowledgement of the original source, for the duration of the public health emergency.

Table 5.

Results of the forward stepwise regression of calculated HCO ₃ ⁻ as dependent variable versus jugular venous values of the physicochemical variables.

Order of entry into regression model dS1	Variable	Partial R 2	Model R 2
1	SIG	0.59	0.59
2	SID3	0.34	0.93
3	Atot	0.04	0.97
4	pCO2	0.01	0.98
Order of entry into regression model DS2
1	SIG	0.42	0.42
2	SID3	0.48	0.90
3	Atot	0.06	0.96
4	pCO2	0.02	0.98

DS1, data set 1; DS2, data set 2; HCO₃ ⁻, bicarbonate; pCO₂, partial carbon dioxide pressure; SID₃, strong ion difference; SIG, strong ion gap; A_tot, total plasma concentration of nonvolatile weak acids. P < .0001 for all the variables' coeffficients.

This article is being made freely available through PubMed Central as part of the COVID-19 public health emergency response. It can be used for unrestricted research re-use and analysis in any form or by any means with acknowledgement of the original source, for the duration of the public health emergency.

Association of Acid–Base Variables with Severity of Depression in Calves with Diarrhea

Plasma concentrations of SIG and AG were significantly different between calves at all 3 clinical scoring categories for attitude, posture, and suckle reflex (P < .0001) (Table 5). Plasma [BE_(efc)] and [HCO₃ ⁻] did not differ significantly between lethargic and stuporous calves, calves in sternal versus lateral recumbency, calves with a weak versus strong suckle reflex, or calves with weak versus absent suckle reflex. SID₃, A_tot, and pCO₂ did not differ significantly among the different clinical scoring categories for attitude, suckle reflex, and posture (Table 6).

Table 6.

Comparison of acid–base variables values for calves with normal, moderate, and severe changes on posture, attitude, and suckle reflex.

Clinical Variable	Attitude			Suckle Reflex			Posture
Acid–Base Variable	Bright (n = 44)	Obtunded (n = 169)	Stuporous (n = 51)	Strong (n = 62)	Weak (n = 110)	Absent (n = 92)	Standing (n = 88)	Sternal (n = 92)	Lateral (n = 8)
H–H
HCO3 −	24.7a	22.3a	16.9b	21.4a	19.3ab	17b	21.7a	18.4b	16.8b
pCO2	51.3a	48.3a	53.2a	49.2a	48.9a	49.3a	50.9a	48a	51.6a
BE(ecf)	−1.9a	−10.5b	−11.4b	−5.9a	−8.6ab	−12.1b	−5.7a	−10b	−12.3b
AG	18a	22.8b	26.3c	19.1a	22.4b	25.5c	19a	23b	26.2c
SID
SID3	42.7a	41a	43a	40.6a	41.8a	42.4a	40.8a	41.5a	42.2a
SIG	−5.3a	−10.8b	−15.8c	−6.3a	−10.5b	−14.3c	−6.2a	−11.2b	−15.2c
Atot	22.2a	22.3a	22a	22.3a	21.8a	22.5a	22.1a	22.3a	21.9a

Data presented as mean values. Different letters within a row indicated a statistically significant difference (P < .001). HCO₃ ⁻, bicarbonate; pCO₂, partial carbon dioxide pressure; BE_(ecf), base excess extracellular fluid; AG, anion gap; SID₃, strong ion difference measured; SIG, strong ion gap; A_tot, total plasma concentration of nonvolatile weak acids.

This article is being made freely available through PubMed Central as part of the COVID-19 public health emergency response. It can be used for unrestricted research re-use and analysis in any form or by any means with acknowledgement of the original source, for the duration of the public health emergency.

Discussion

Use of the Henderson–Hasselbalch and physicochemical approach to evaluate acid–base status indicated that acidemia was the most common acid–base disorder in the calves with diarrhea included in this study. Acidemia was predominant because of an increase in the strong UAs concentration (increased SIG) and changes in the concentration of the measured strong ion concentration (SID₃). Increase in the total plasma protein concentration resulting in weak acids acidosis only had a minor contribution to the acidemia in this group of calves. Stewart's major contribution to the understanding of clinical acid–base physiology was his proposal that plasma [H⁺] is determined by 3 independent factors: pCO₂, [SID], and A_tot ⁻.9 SID has 2 components: measured SID and unmeasured SID (SID_um).7 The simplified strong ion approach quantifies the unmeasured strong ions by calculating the SIG from values of plasma or serum concentration of strong anions and strong cations, the total protein or albumin concentration, and experimentally determined values of A_tot and pK_a in calf plasma, and can be used interchangeably for SID_um.7, 10, 32 Theoretically, [SID] can be altered in several ways to produce metabolic acidosis, such as a reduction in [Na⁺], an increase in [Cl⁻], endogenous production of strong anions (l‐lactic acidosis), absorption of strong anions (d‐lactic acidosis), and impaired excretion of unmeasured anions (uremic acids).7, 33, 34 Plasma electrolyte disorders including hyponatremia, in the presence of normochloremia or hyperchloremia, were observed in some of the calves in these data sets and were the likely cause of the decreased of the SID₃. This finding confirms the important role of hyponatremia, attributed to excessive loss of sodium into the gastrointestinal tract and decreased milk intake because of anorexia,6, 35, 36 in producing acidemia and strong ion acidosis in calves with diarrhea.7

In this retrospective study in calves with diarrhea the major contributor to the changes in plasma pH and HCO₃ ⁻ was an increase in unmeasured anions. Combined the SIG and SID₃ accounted for 77 and 93% of the changes in the measured venous pH and calculated HCO₃ ⁻ in calves of DS1, respectively, whereas SIG and SID₃ accounted for 64 and 93% of the changes in the measured venous pH and calculated HCO₃ ⁻ in calves of DS2, respectively. Similar results were reported previously in sick calves in which SID_um and SID₃ accounted for 58–75% of the changes of the calculated pH; it was presumed that [d‐lac⁻] was the predominant unmeasured anion.7 Hyper‐d‐lactatemia occurs frequently in diarrheic calves.21, 22, 23[d‐lac⁻] acid is a major contributor to anion gap acidosis, accounting for approximately 64% of the total increase in organic acids in plasma.21 Malabsorption of nutrients in the small intestine of patients with short bowel syndrome and subsequent fermentation by colonic bacteria result in d‐lactic acidosis in humans.37 d‐lactate concentrations are significantly higher in rumen content, feces, serum, and urine in diarrheic calves with acidemia as compared to normal calves.22, 38 Our findings support the clinically useful application of SID theory in the calculation of the SIG to quantify the unmeasured strong anion concentration such as [d‐lac⁻] and [l‐lac⁻] in calves with diarrhea, and the calculation of SID₃ to quantify the effects of [Na⁺] and [Cl⁻] on acid–base status.39

Strong ion gap may also be reduced by an increase in the strong anions such as l‐lactate. There is a strong evidence that neonatal diarrhea predisposes calves to septicemia.3, 28, 40, 41, 42 One of the most important consequences of the septicemia is the compromise of organ perfusion because of the arterial hypotension. Furthermore, endotoxin impairs oxygen extraction by tissues.43 As a result, anaerobic glycolysis, combined with direct inhibition of pyruvate dehydrogenase by endotoxin, generates severe l‐lactic acidosis.44, 45 Contradictory results regarding the contribution of the concentration of [l‐lac⁻] to the acidemia in diarrheic calves have been reported. In some studies, the contribution of [l‐lac⁻] to the metabolic acidosis in calves with diarrhea has been attributed to the severity of the dehydration and the age of the calf,19, 22 whereas other studies have shown that [l‐lac⁻] was a minor contributor the acidemia in diarrheic or severely dehydrated calves.22, 26, 46 The results of this study revealed that a high proportion of the calves were dehydrated and had strong ion metabolic acidosis, and that one‐third were considered to be septicemic, 70% of which did not survive. These findings suggest that some of the calves were in a progressive or irreversible phase of hypovolemic or septic shock or both where hypotension and hypoperfusion develop, leading to an increase in plasma [l‐lac⁻], which could have contributed in part to the increased concentration of the unmeasured strong ions detected by the calculation of the SIG.47 l‐lactate concentrations were not measured in the calves included in this study.

Based on our results, we speculate that the causes for the strong ion acidosis in this group of calves include (1) increased loss of fluid with high [Na⁺] and low [Cl⁻] through the feces; (2) increased plasma [l‐lac⁻] caused by dehydration and septicemia; and (3) absorption of [d‐lac⁻] from the gastrointestinal tract. These findings emphasize the importance of identifying the presence of strong UAs by calculating SIG when applying the physicochemical approach. Additionally, this study highlights the role of strong ions in the acid–base status of calves with diarrhea, and indicates that replacement fluid therapy in animals with diarrhea should be aimed at resolving volume deficits, improving peripheral perfusion, and correcting [Na⁺] and [Cl⁻] imbalances.48, 49

This retrospective study revealed that clinical scores for posture, attitude, and suckle reflex were influenced primarily by the magnitude of the increased plasma concentration of unmeasured strong anions (increased SIG and AG). Early studies demonstrated that demeanor and posture in calves with diarrhea were influenced by the degree of metabolic acidosis,16 and that base deficit had the strongest correlation with alteration of demeanor,16, 24 behavior,24 and suckle reflex.20, 24 Conversely, a recent study failed to demonstrate that severe, uncomplicated, hyperchloremic acidosis was associated with the demeanor abnormalities accompanying neonatal calves with diarrhea.25 Calves with naturally acquired diarrhea have variations in behavior, posture, and palpebral reflex, but not in the suckle reflex, that could be better explained by increases in serum [d‐lac⁻] than by decreases in base excess.24 Both natural and experimentally induced hyper‐d‐lactatemia in calves produced an impaired palpebral reflex, somnolence, and a staggering gait, but no impairment of the suckling reflex.18, 50 It has been suggested than metabolic acidosis without appreciable increase in [d‐lac⁻] has only a minor influence on posture and behavior of calves with diarrhea51; however, posture and especially demeanor normalized after correction of the acidemia with sodium bicarbonate despite a persistent increase in [d‐lac⁻] concentration in calves with diarrhea and metabolic acidosis.51 Moreover, in 1 study of calves with naturally acquired diarrhea and strong ion acidosis that were treated with intravenous solution of sodium, behavior was normalized despite of the persistent increases in [d‐lac⁻] above 3 mmol/L.30 In this retrospective study SID₃ and pCO₂ did not correlate with changes in attitude, suckle reflex and posture; additionally, A_tot was not significantly correlated with attitude and suckle response although it did correlate with posture. This supports the hypothesis that unmeasured strong anions including [d‐lac⁻], [l‐lac⁻], and uremic anions can be the major cause for deteriorating demeanor parameters, and suggest that fluid volume replacement to re‐establish tissue perfusion and diuresis with polyionic crystalloid alkalinizing solution is crucial for the treatment of calves with diarrhea.

There were several limitations to our findings, most notably the retrospective design, as well as the selected population being referred to the VTH, which would tend to be biased toward sicker patients. Secondly, the data were collected from 2 different retrospective data sets, which can increase the limitations and inaccuracies of the retrospective studies. A further validation of these results through a prospective study is warranted. Finally, 1 potential limitation of this study was the use of 2 different blood gas analyzers with 2 different values of S and pK₁′ for calculation of plasma HCO₃ ⁻ from pCO₂ and pH values.52 Assigned values for S and pK₁′ for DS1 were 0.0306 mmol/L/mmHg and 6.099 mmol/L, respectively; assigned values for S and pK₁′ for DS2 were 0.0307 mmol/L/mmHg and 6.095 mmol/L, respectively. However, the rearrangement of the H–H equation as pH = [(−log₁₀K₁ − log₁₀(S)] + log₁₀ (HCO₃ ⁻/pCO₂) permitted the combination of the 2 constants into one, expressed as [pK₁ − log₁₀(S)]. Substituting S and pK₁′ for the assigned values for each blood gas resulted in [6.099 + 1.514 = 7.613] for DS1 and [6.095 + 1.513 = 7.608] for DS2; these values differ by 0.066%. A difference of 0.066% between DS1 and DS2 for the calculation of plasma concentration of HCO₃ ⁻ seemed to be of little importance for the purposes of this study.

In conclusion, the results of this study revealed that UAs and strong UAs concentrations calculated using AG and SIG, respectively, were the primary causes of acidemia in calves with diarrhea. They also had a greater influence on clinical scores of attitude, suckle reflex, and posture, as compared to BE_(ecf) and [HCO₃ ⁻]. These findings emphasize the importance of the calculation of the strong ion gap or anion gap when evaluating acid–base abnormalities in diarrheic calves.