Bicarbonate Values for Healthy Residents Living in Cities Above 1500 Meters of Altitude: A Theoretical Model and Systematic Review

Abstract

Ramirez-Sandoval, Juan C., Maria F. Castilla-Peón, José Gotés-Palazuelos, Juan C. Vázquez-García, Michael P. Wagner, Carlos A. Merelo-Arias, Olynka Vega-Vega, Rodolfo Rincón-Pedrero, and Ricardo Correa-Rotter. Bicarbonate values for healthy residents living in cities above 1500 m of altitude: a theoretical model and systematic review. High Alt Med Biol. 17:85–92, 2016.—Plasma bicarbonate (HCO₃⁻) concentration is the main value used to assess the metabolic component of the acid–base status. There is limited information regarding plasma HCO₃⁻ values adjusted for altitude for people living in cities at high altitude defined as 1500 m (4921 ft) or more above sea level. Our aim was to estimate the plasma HCO₃⁻ concentration in residents of cities at these altitudes using a theoretical model and compare these values with HCO₃⁻ values found on a systematic review, and with those venous CO₂ values obtained in a sample of 633 healthy individuals living at an altitude of 2240 m (7350 ft). We calculated the PCO₂ using linear regression models and calculated plasma HCO₃⁻ according to the Henderson–Hasselbalch equation. Results show that HCO₃⁻ concentration falls as the altitude of the cities increase. For each 1000 m of altitude above sea level, HCO₃⁻ decreases to 0.55 and 1.5 mEq/L in subjects living at sea level with acute exposure to altitude and in subjects acclimatized to altitude, respectively. Estimated HCO₃⁻ values from the theoretical model were not different to HCO₃⁻ values found in publications of a systematic review or with venous total CO₂ measurements in our sample. Altitude has to be taken into consideration in the calculation of HCO₃⁻ concentrations in cities above 1500 m to avoid an overdiagnosis of acid–base disorders in a given individual.

Introduction

Around the world, millions of people living at altitudes above 1500 m (4921 ft) present with medical conditions that can be associated with acid–base disorders. Plasma bicarbonate (HCO₃⁻) concentration is the main value traditionally used to assess the metabolic component of the acid–base status in a patient (Adrogué et al., 2009). Despite its widespread use, there is a lack of medical information regarding the effect of altitude in the standard HCO₃⁻ values measured by arterial blood gases (ABGs) in patients living at elevated altitudes.

High altitude can be defined as >1500 m (4921 ft) over the sea level, given that at this altitude a state of hyperventilation occurs due to the fall in partial pressure of inspired oxygen (PO₂) decreasing the partial pressure of carbon dioxide (PCO₂) as the minute ventilation increases (Bernardi et al., 2006; Paralikar.and Paralikar, 2010; Hackett and Roach, 2012; Goldfarb-Rumyantzev and Alper, 2014). According to the Henderson–Hasselbalch equation, the extracellular fluid alkalinization induced by the drop of plasma PCO₂ is compensated by a fall in the concentration of HCO₃⁻. Moreover, it has been shown that increasing altitude produces a proportional elevation in the renal excretion of HCO₃⁻, which occurs proportionate to elevation even at low and moderate altitude exposure (Ge et al., 2006).

Some studies have shown that the kidney is able to modify the degree of HCO₃⁻ tubular reabsorption guided by PCO₂ changes, affecting its urinary excretion, to balance the chronic hypocapnic response in residents of high altitude cities (Monge et al., 1964). Nonetheless, information on the normal values of plasma HCO₃⁻ in healthy residents at high altitude is minimal.

The vast majority of clinical laboratories calculate HCO₃⁻ concentrations using the Henderson–Hasselbalch equation: cHCO₃⁻ = pH + log (PCO₂ × 0.0307) − 6.095 (Ehrmeyer et al., 1997). The caveat of this procedure is that PCO₂ is affected by altitude exposure; therefore, reference values that have been obtained from subjects at sea level or very high altitude (Crapo et al., 1999) may provide inaccurate results in residents of low to moderate altitude cities.

Due to the small amount of studies that address the reference values of plasma HCO₃⁻ according to altitude and given the relevance of the molecule in the clinical assessment of acid–base status, it is necessary to generate altitude-expected HCO₃⁻ plasma concentrations. The aim of this analysis is to calculate the reference concentrations of plasma HCO₃⁻ in patients living at an altitude of 1500 m or more meters using a theoretical model and compare these estimated HCO₃⁻ values with HCO₃⁻ values from residents of cities at >1500 m (4921 ft) obtained from a literature search and with those venous CO₂ values obtained in a sample of healthy residents living at an altitude of 2240 m (7350 ft).

Methods

For the present analysis, we defined high altitude as 1500 m above sea level or more, in agreement with other authors and considering that at that altitude, decreased exercise performance and/or significant declines in PO₂ to levels of 55 to 75 mmHg can be observed (Grocott et al., 2009).

HCO₃⁻ concentrations were calculated using the Henderson–Hasselbalch equation (cHCO₃⁻ = pH + log [PCO₂ _× 0.0307] − 6.095) as described in the Clinical Laboratory Standards document (Ehrmeyer et al., 1997). Also, PCO₂ values were calculated using two linear regression models for acute exposure to altitude (PCO₂ = 37.78 − 0.908A; where A is the altitude over the sea level in km), and after acclimatization (PCO₂ = 38.3 − 2.5A), PCO₂ is shown in mmHg. These equations are calculated from a linear regression model with two points, using normal PCO₂ measurements in Mexico City (2240 m) and sea level showed in a previous study (Pérez Padilla and Vázquez García, 2000). “Acute exposure” refers to values from people who ascend from sea level to high altitudes in less than 1 week based on previous observations (Purkayastha et al., 1995), whereas “acclimatized residents” refers to values from people living many years in the referred city. Demographic indices were taken from the United Nations Statistics Division (United Nations Demographic Yearbook, 2013).

We compared the estimated HCO₃⁻ values with two sources: (1) the historic HCO₃⁻ values reported in healthy subjects living at >1500 m above sea level or more and (2) the venous CO₂ values of 633 healthy adult permanent residents of Mexico City (altitude 2240 m [7350 ft] over sea level).

The HCO₃⁻ concentration at different high altitude locations were obtained by systematic review of literature. We searched in MEDLINE (January 1, 1950 to December 31, 2015) and LILACS (January 1, 1950 to December 31, 2015). The terms, “Blood Gas Analysis”, “blood gas”, “Acid–Base Equilibrium”, “CO2”, “PCO2”, “HCO3,” or “bicarbonate” combined with, “altitude”, ”Mexico”, “Peru”, “Colombia”, “Bolivia”, “Afghanistan”, “Ethiopia”, “Africa”, or “Guatemala” as free language terms and medical subject headings when applicable, were used in the search strategy. Included studies were those with blood gas values reported in healthy or apparently healthy adults living at >1500 m or more above sea level for years (“Acclimatized residents”) or in nonacclimatized adults (“Acute exposure”) exposed to high altitudes. The excluded studies were: (1) studies not having bicarbonate values, (2) studies based on population of less than 16 years of age, or with (3) comorbidities (obesity, chronic obstructive pulmonary disease, smokers, cardiovascular disease, etc.), and (4) studies with a small sample size, which may compromise the power and external validity of the findings (e.g., sample <10). Only English and Spanish articles were included. Duplicated citations were discarded and two independent researchers (M.F.C.P. and C.A.M.A.) eliminated irrelevant titles and abstracts. Remaining citations were extensively reviewed by both researchers. Disagreements were discussed with a third reviewer (J.C.R.S.) to reach consensus. References from relevant original articles and reviews were screened for additional citations. We independently extracted data of selected studies.

Given the reduced number of studies in the literature, we decided to compare our theoretical values with venous CO₂ samples of 633 residents from the Salmex cohort. Generally, the HCO₃⁻ comprises about 95% of the total CO₂ content; thus we use this measurement as an estimator of serum HCO₃⁻ (Centor, 1990). The Salmex cohort consists of adult male and female workers from the National Institute of Medical Sciences and Nutrition Salvador Zubirán, in Mexico City. Exclusion criteria were: history of congestive heart failure, renal or hepatic disease, intestinal resection, diuretic initiation in the previous 5 days, active infection, pregnancy, lactation, currently smoking (33/666), and previous known or de novo diagnosis of hypertension, chronic kidney disease, or diabetes mellitus. All participants received a detailed explanation of the tasks to be undertaken and appropriate containers for a 24-hour urine collection. A later appointment was scheduled for complete medical history, physical examination, nutritional and body composition evaluation, and laboratory work-up, which included delivery of the 24-hour urine collection and a fasting blood sample.

Venous blood samples were drawn without use of a tourniquet. All samples were sent immediately to the laboratory and analyzed using ion-selective electrode technology developed for monitoring CO₂ in human blood (Beckman Synchron CX^® 5, Fullerton, CA). The variation coefficient of the test was 3%.

All procedures were conducted according to the Declaration of Helsinki guidelines and approved by the Institutional Biomedical Research Review Board (number 09/191). Written informed consent was obtained from all subjects before baseline data and sample collection.

CO₂ results are presented as mean standard deviation for normally distributed data or median interquartile range (IQR) for non-normally distributed data. The Mann–Whitney U test was used for comparison of CO₂ according to gender. A linear regression model fitted to the change of HCO₃⁻ with altitude was used to compare the slopes between HCO₃⁻ reported in literature and the theoretical model. Bland–Altman analysis was performed to evaluate the agreement between HCO₃⁻ estimated and HCO₃⁻ described in literature. All the documented data were analyzed using the SPSS 21 program (Chicago, IL) and GraphPad Prism 6 (San Diego, CA). The results were considered statistically significant when p < 0.05.

Results

As seen in Tables 1 –3, there is a decrease in HCO₃⁻ concentration as the altitude increases, with the lowest value measured from the highest city in our list. All mean values of PCO₂ and HCO₃⁻ in acclimatized residents were below 36 mmHg and 21 mEq/L, which are less than reference values reported in arterial samples taken in persons living at sea level (Emmet, 2015). The fall in HCO₃⁻ is proportional to the drop in PCO₂.

Table 1.

Estimated Values for PCO₂ and HCO₃ ⁻ in Cities >1500 m Assuming a pH of 7.4 (>4921 ft)

		Acute exposure ^a		Acclimatized residents
Altitude (m)	Altitude (ft)	PCO₂ (mmHg)	HCO₃⁻ (mEq/L)	PCO₂ (mmHg), ref. value 35–45 ^b	HCO₃⁻ (mEq/L), ref. value 21–27 ^b
1500	4921	36.4	22.6	34.6	20.7
1600	5249	36.3	22.5	34.3	20.5
1700	5577	36.2	22.5	34.1	20.4
1800	5905	36.1	22.4	33.8	20.2
1900	6234	36.1	22.3	33.6	20.1
2000	6562	36.0	22.3	33.3	19.9
2100	6890	35.9	22.2	33.1	19.8
2200	7218	35.8	22.2	32.8	19.6
2300	7546	35.7	22.1	32.6	19.5
2400	7874	35.6	22.1	32.3	19.3
2500	8202	35.5	22.0	32.1	19.2
2600	8530	35.4	22.0	31.8	19.0
2700	8858	35.3	21.9	31.6	18.9
2800	9186	35.2	21.9	31.3	18.7
2900	9514	35.1	21.8	31.1	18.6
3000	9842	35.1	21.7	30.8	18.4
3100	10,170	35.0	21.7	30.6	18.3
3200	10,499	34.9	21.6	30.3	18.1
3300	10,827	34.8	21.6	30.1	18.0
3400	11,155	34.7	21.5	29.8	17.8
3500	11,483	34.6	21.5	29.6	17.7
3600	11,811	34.5	21.4	29.3	17.5
3700	12,139	34.4	21.4	29.1	17.4
3800	12,467	34.3	21.3	28.8	17.2
3900	12,795	34.2	21.3	28.6	17.1
4000	13,123	34.1	21.2	28.3	16.9
5000	16,404	33.1	19.8	25.6	15.4
6000	19,685	32.3	19.4	23.3	13.9
7000	22,966	31.4	18.8	20.8	12.5
8000	26,247	30.5	18.3	18.3	11.0
8400^c	27,559	30.2	18.0	17.3	10.4

Acute exposure refers to values from people who rapidly ascend from sea level, whereas acclimatized residents refer to values from people living at the city referred.

Reference values (Theodore, 2015).

Maximal altitude with HCO₃⁻ values reported (Grocott et al., 2009).

Table 2.

Estimated Values for PCO₂ and HCO₃ ⁻ in Capital Cities >1500 m (>5000 ft) Assuming a pH of 7.4

				Acute exposure ^a		Acclimatized residents
City	Country	Altitude (m)	Altitude (ft)	PCO₂ (mmHg)	HCO₃⁻ (mEq/L)	PCO₂ (mmHg), ref. value 35–45 ^b	HCO₃⁻ (mEq/L), ref. value 21–27 ^b
Quito	Ecuador	2850	9350	35.2	21.1	31.2	18.7
Sucre	Bolivia	2750	9022	35.3	21.1	31.4	18.8
Thimphu	Bhutan	2648	8688	35.4	21.2	31.7	19.0
Bogota	Colombia	2625	8612	35.4	21.2	31.7	19.0
Addis Abeba	Ethiopia	2355	7726	35.6	21.3	32.4	19.4
Asmara	Eritrea	2325	7628	35.7	21.4	32.5	19.4
Sana'a	Yemen	2250	7382	35.7	21.4	32.7	19.6
Mexico City	Mexico	2240	7350	35.7	21.4	32.7	19.6
Nairobi	Kenya	1795	5889	36.2	21.6	33.8	20.2
Kabul	Afghanistan	1790	5873	36.2	21.6	33.8	20.2
Windhoek	Namibia	1721	5646	36.2	21.7	34.0	20.4
Maseru	Lesotho	1673	5489	36.3	21.7	34.1	20.4
Kigali	Rwanda	1567	5141	36.4	21.8	34.4	20.6
Lima	Peru	1550^c	5080	36.4	21.8	34.4	20.6
Guatemala City	Guatemala	1529	5016	36.4	21.8	34.5	20.6

Acute exposure refers to values from people who rapidly ascend from sea level, whereas acclimatized residents refer to values from people living at the city referred.

Reference values (Theodore, 2015).

Lima is located on mostly flat terrain, but the city slopes into mountain slopes located as high as 1550 m (5090 ft) above sea level.

Table 3.

Estimated Values for PCO₂ and HCO₃ ⁻ in Other Cities >1500 m (Population Greater Than 1,000,000 Inhabitants) Assuming a pH of 7.4

				Acute exposure		Acclimatized residents
City	Country	Altitude (m)	Altitude (ft)	PCO₂ (mmHg)	HCO₃⁻ (mEq/L)	PCO₂ (mmHg)	HCO₃⁻ (mEq/L)
El Alto	Bolivia	4070	13,352	34.0	20.4	27.9	16.7
La Paz	Bolivia	3640	11,942	34.5	20.6	29.2	17.5
Arequipa	Peru	2335	7660	35.7	21.3	32.5	19.4
Puebla	Mexico	2147	7044	35.8	21.4	32.9	19.7
Johannesburg	South Africa	1750	5741	36.2	21.7	33.9	20.3
Denver	USA	1610	5282	36.3	21.7	34.3	20.5
Medellin	Colombia	1495	4905	36.4	21.8	34.6	20.7

According to the theoretical model, assuming an HCO₃⁻ value of 22.9 mEq/L at sea level and a pH of 7.40, for each 1000 m of altitude above sea level, there is a decrease of 0.55 mEq/L in people with an acute exposure to elevation (95% CI 1.45–1.51). In acclimatized people, the decrease is 1.5 mEq/L per each 1000 m (95% CI 5.2–5.8) (Fig. 1).

FIG. 1.

Bicarbonate values expected for altitude.

HCO₃⁻ in “Acute exposure” to altitude = 22.89 − (0.55 ± 0.01 × altitude in km).

HCO₃⁻ in “Acclimatized” = 22.89 − (1.47 ± 0.01 × altitude in km).

Estimated HCO₃⁻ values were compared with those obtained in a systematic review of the literature. Our research identified 965 studies of which 752 did not meet inclusion criteria. The reasons for exclusion were no HCO₃⁻ values (169), incomplete information (22), subjects with less than 16 years of age (6), and inclusion of patients with pulmonary or cardiovascular diseases (3). A total of 13 studies were considered for further analysis. Data in “Acclimatized” subjects are reported separately from “Acute exposure” subjects. Studies are summarized in Tables 4 and 5. Gaps were identified in the published literature: small samples, poor definition of populations, great variability in values of HCO₃⁻, and heterogeneity in protocols and methods. One small study, excluded in Table 4, was relevant as it is the only field measurement of ABGs at 8400 m (“the balcony,” Mount Everest), and reported mean concentrations of HCO₃⁻ 10.8 mEq/L in four well-acclimatized subjects, similar to our estimated HCO₃⁻ of 10.3 mEq/L (Table 1) (Grocott et al., 2009).

Table 4.

Values in Literature for PCO₂ and HCO₃ ⁻ in Residents Living in Cities >1500 m (“Acclimatized Residents”)

Altitude, m (ft)	Reference	Number of subjects	Age, years (±SD) or range	PCO₂, mmHg	HCO₃⁻, mEq/L	Theoretical value HCO₃⁻, mEq/L	Comments
1547 (5075)	Rico et al. (2001)	30	67 ± 4.3	32.6 ± 2.8	21.2 ± 1.3	20.6	Arterial blood
1768 (5801)	Gahutu et al. (2005)	ND	20–37	34.7 ± 3.4 (male) 31.8 ± 3.1 (female)	21.6 ± 2.2 (male) 20.4 ± 2.6 (female)	20.3	Arterial blood
2144–2240 (7034–7349)	Rico et al. (2001)	60	67.4 ± 6.0	32.6 ± 3.7	22.3 ± 2.5	19.5 to 19.7	Arterial blood
2238 (7343)	Rico et al. (1998)	102	60–70	ND	21.5 ± 2.8	19.6	Arterial blood
		102	71–80	ND	22 ± 3	19.6	Arterial blood
		102	>80	ND	20.4 ± 4.13	19.6	Arterial blood
2625 (8612)	Acevedo and Solarte (1984)	25	22.1 ± 3.4	29.5 ± 2.14	16.9 ± 1.5	19	Arterial blood
2640 (8661)	González-García et al. (2004)	16	ND	30.3 ± 2.4	20.2 ± 1.7	19.0	Arterial blood
3100 (10,171)	Forster et al. (1969)	10	27 ± 0.5	36 ± 0.49	22.7 ± 0.6	18.3	Arterial blood
4300 (14,108)	Hansen et al. (1967)	8	ND	26.2	15.4 ± ND	16.5	Arterial blood
4700 (15,420)	Ge et al. (1994)	31 (17 Tibetan and 14 Han native)	29.4 ± 1.4	25.8 ± 0.9	16.9 ± 0.6^a	15.9	Capillary blood (ear)

Study of Ge et al. (1994).

ND, no data; SD, standard deviation.

Table 5.

Values in Literature for PCO₂ and HCO₃⁻ in People Living at Sea Level and Traveling to Cities >1500 m with Less Than 1 Week of Acclimatization (“Acute Exposure”)

Altitude, m (ft)	Reference	Number of subjects	Age, years (±SD), or range	PCO₂, mmHg	HCO₃⁻, mEq/L	Theoretical value HCO₃⁻, mEq/L	Comments
3200 (10,499)	Dempsey et al. (1978)	21	21–48	33.9	25.6	21.3	Arterial blood
3500 (11,483)	Purkayastha et al. (1995)	60	24.3 ± 3.1	35.5–32.5	21.9–22.0	20.7	Arterial blood taken after 5 days at 3500 m
4300 (14,108)	Hansen et al. (1967)	8	ND	28.1	17.9	20.3	Arterial blood
4900 (16,076)	Böning et al. (2001)	12	30.8 ± 1.9	34.4 ± 1.1	22.0 ± 0.5	20.0	Capillary blood (ear)
5050 (16,568)	Fan et al. (2012)	12	31 ± 9	29 ± 3	21.3 ± 2.4	19.8	Arterial blood
3440 (11,286)	Wright et al. (2004)	20	24–59	27.5 ± 2.4	23.3 ± 0.7	20.7	Arterialized capillary samples
4120 (13,517)	Wright et al. (2004)	20	24–59	24.7 ± 2.7	22.3 ± 1.6	20.4	Arterialized capillary samples
5200 (17,060)	Wright et al. (2004)	20	24–59	23.7 ± 6.2	20.3 ± 1.5	19.8	Arterialized capillary samples

We estimated a linear regression model including this last study.

The model calculated from the systematic revision was:

HCO₃⁻ in “acclimatized resident” (obtained from literature) = 24.43 − (1.64 ± 0.26 × altitude in km).

HCO₃⁻ in “acute exposure” (obtained from literature) = 32.81 − (2.54 ± 0.83 × altitude in km).

The slopes for change in HCO₃⁻ according to altitude were not statistically significant when we compare the theoretical model and the reported values of the literature (Figs. 2 and 3). Bland–Altman plot using differences between HCO₃⁻ estimated and HCO₃⁻ reported in literature revealed a bias of 1.05 mEq/L (standard deviation 1.6 mEq/L) and the limits of agreement were −2.097 and 4.203 mEq/L, with a span of 6.3 mEq/L. Out of 20 values, 19 (95%) were within the limits of agreement (Fig. 4).

FIG. 2.

Linear regression model calculated from HCO₃⁻ values reported in literature and HCO₃⁻ estimated according to our theoretical model in acclimatized residents. The comparison between the two slopes was not significant (p = 0.09).

FIG. 3.

Linear regression model calculated from HCO₃⁻ values reported in literature and HCO₃⁻ estimated according to our theoretical model in nonacclimatized or acute exposure subjects. The comparison between the two slopes was not significant (p = 0.19).

FIG. 4.

Bland–Altman plot of average of HCO₃⁻ estimated and HCO₃⁻ reported versus the difference between the two methods in mEq/L. LOA, limit of agreement.

To compare HCO₃⁻ estimated values with a large sample of subjects, we measured CO₂ values from venous samples of 633 healthy residents at 2240 m (7350 ft), 433 (68%) of them were female and the mean age was 38 years ± 10 years. The total CO₂ content had a non-normal distribution, the median CO₂ value was 22.3 mEq/L (IQR 20.7–24.9). These values were slightly superior to the HCO₃⁻ predicted value of 19.1 mEq/L in acclimatized residents of Mexico City. Median total CO₂ levels were 23.4 and 22.0 in males and females, respectively (p < 0.0001) (Fig. 5). Of all CO₂ values, 95% were between 19.6–26.9 mEq/L. Using a reference value between 22–30 mEq/L for total venous CO₂, 277 healthy patients (44%) had a venous CO₂ less than 22 mEq/L and could have been subjected to unnecessary tests.

FIG. 5.

Horizontal mirrored histogram with total venous CO₂ in 633 residents living at 2240 m (7350 ft) according to gender. Median total CO₂ levels were 23.4 and 22.0 in males and females, respectively, and 277 values (44%) were less than 22 mEq/L.

Discussion

We hereby present the calculated mean HCO₃⁻ values in healthy residents living at different altitudes according to the Henderson–Hasselbalch equation. These values are consistent with data obtained in a sample of healthy residents and with other data summarized in 13 publications. Our equation, utilized to obtain bicarbonate values in acclimatized patients, is similar to another proposed by Zubieta-Calleja in which HCO₃⁻ = 24.32 − (1.8 × altitude in km) (Zubieta-Calleja et al., 2011). Using Bland–Altman analysis, the difference in means (bias) between HCO₃⁻ values estimated and reported in literature was almost 1 mEq/L, a difference clinically unimportant. In general, all healthy acclimatized residents living at 1500 m or more above sea level will show lower PCO₂ and HCO₃⁻ compared with residents at sea level under physiological circumstances and with a normal pH (7.40 ± 0.05).

The plasma bicarbonate concentration [HCO₃-] is widely used to assess the acid–base status of patients. Using the traditional method based in the use of bicarbonate to interpret metabolic disturbances (Rose and Post, 2001), a healthy person living at a specific altitude over sea level may be subject to unnecessary diagnostic procedures to find a medical cause of a mixed acid–base disorder (respiratory alkalosis and metabolic acidosis with normal anion gap), or, according to other authors, compensated respiratory alkalosis, other than considering a healthy physiological adaptation to altitude. Although it is possible that by using Stewart's approach (Kurtz et al., 2008) this error can be avoided, one physician using the traditional bicarbonate-based method can overdiagnose acid–base disorders in people living at or visiting high altitudes with healthy physiological adaptation.

In medicine, some clinical guidelines take into account a specific HCO₃⁻ value either for diagnosis or as a goal of treatment, without emphasizing the effects of geographic altitude in their recommendations. For example, there is a consensus to recommend that patients with chronic kidney disease and with HCO₃⁻ <22 mEq/L use alkali therapy, including NaHCO₃⁻ (Loniewski and Wesson, 2014); in ketoacidosis, specific HCO₃⁻ values have been proposed for classification of severity and for goals of treatment (Nyenwe and Kitabchi, 2011) and low HCO₃⁻ values could be related to overdiagnosis of entities such as renal tubular acidosis in cities with high altitude (Muñoz-Arizpe et al., 2012). None of these recommendations takes into account altitude over sea level of the tested individual in their analysis.

The information available about the bicarbonate values in persons at altitude is scarce. The majority of cities that we have listed belong to low and lower middle income countries, with lower medical publication rates than countries with higher incomes (Akre et al., 2011). Surprisingly, in some reviews and articles in those countries, their recommendations to diagnose acid–base disorders ignore the effects of altitude (Secretaría de salud, 2010; Márquez-González et al., 2012).

The agreement between measured total venous CO₂ and HCO₃⁻ and whether both can be used interchangeably has long been discussed (Chittamma and Vanavanan, 2008; Kumar and Karon, 2008). Our results of venous CO₂ levels in healthy residents of altitude may be different compared with the real values of HCO₃⁻ obtained from venous and ABGs (Lolekha et al., 2003) especially when HCO₃⁻ value is equal to or less than 20 mmol/L (Nasir et al., 2010). Nevertheless, low CO₂ values were observed in this sample, in some cases, lower than 18 mEq/L.

Our study has limitations. The values described in all tables must be taken with caution as they represent theoretical estimates and may vary from true values. The small sample of the studies in the literature can be a source of error. Sex is not considered for reference values of HCO₃⁻ and women may have lower levels of HCO₃⁻ according the CO₂ value results. Calculations started from the estimated barometric pressure derived from the altitude. However, actual barometric pressures may differ due to geographical and atmospheric conditions. Our HCO₃⁻ values are still open for discussion and have to be tested in every city with high altitude. We expect that our observation may favor other groups to perform specific investigations to better understand acid–base physiology and document normal values for individuals living at different altitudes.

In conclusion, HCO₃⁻ values need to be interpreted according to altitude above sea level as well as to an individual's time of exposure to an altitude. Physicians that attend to patients living at elevated cities have to be aware of this and remain careful in assessing acid–base disorders when using algorithms recommended for people that live at sea level unless specific, corrected, formula-generated values are validated for this purpose.

Footnotes

Acknowledgment

The study was supported by a nonrestricted research grant from Instituto Danone, Mexico.

Author Disclosure Statement

No competing financial interests exist.

References

Acevedo

, and Solarte

. (1984). Gasometria arterial en adultos jóvenes a nivel de Bogota. Acta Med Col, 9:7–13.

Adrogué

, Gennari

, Galla

, and Madias

. (2009). Assessing acid-base disorders. Kidney Int, 76:1239–1247.

Akre

, Barone-Adesi

, Pettersson

, Pearce

, Merletti

, and Richiardi

. (2011). Differences in citation rates by country of origin for papers published in top-ranked medical journals: do they reflect inequalities in access to publication?. J Epidemiol Commun Health, 65:119–123.

Bernardi

, Schneider

, Pomidori

, Paolucci

, and Cogo

. (2006). Hypoxic ventilator response in successful extreme altitude climbers. Eur Respir J, 27:165–171.

Böning

, Maassen

, Thomas

, and Steinacker

. (2001). Extracellular pH defense against lactic acid in normoxia and hypoxia before and after a Himalayan expedition. Eur J Appl Physiol, 84:78–86.

Centor

. (1990). Serum total carbon dioxide, chapter 196. In: Clinical Methods: The History, Physical, and Laboratory Examinations. Walker

, Hall

, and Hurst

, eds. Butterworths, Boston. pp. 1–5.

Chittamma

, and Vanavanan

. (2008). Comparative study of calculated and measured total carbon dioxide. Clin Chem Lab Med, 46:15–17.

Crapo

, Jensen

, Hegewald

, and Tashkin

. (1999). Arterial blood gas reference values for sea level and an altitude of 1,400 meters. Am J Respir Crit Care Med, 160:1525–1531.

Dempsey

, Forster

, Chosy

, Hanson

, and Reddan

. (1978). Regulation of CSF[HCO3-] during long-term hypoxic hypocapnia in man. J Appl Physiol Respir Environ Exerc Physiol, 44:175–182.

10.

Ehrmeyer

, Burnett

, and Chatburn

. (1997). Fractional Oxyhemoglobin, Oxygen Content and Saturation, and Related Quantities in Blood: Terminology, Measurement, and Reporting. National Committee for Clinical Laboratory Standards, Wayne, PA. NCCLS document C25-A.Vol. 12 (11).

11.

Emmet

. (2015). Simple and mixed acid-base disorders. In: UpToDate. Sterns

and Forman

, eds. UpToDate, Waltham, MA (accessed on June 15, 2015).

12.

Fan

, Burgess

, Thomas

, Lucas

, Cotter

, Kayser

, et al. (2012). Effects of acetazolamide on cerebrovascular function and breathing stability at 5050 m. J Physiol, 590:1213–1225.

13.

Forster

, Dempsey

, Birnbaum

, Reddan

, Thoden

, Grover

, et al. (1969). Comparison of ventilatory responses to hypoxic and hypercapnic stimuli in altitude-sojourning lowlanders, lowlanders residing at altitude and native altitude residents. Fed Proc, 28:1274–1279.

14.

Gahutu

, Wane

, d'Arc Uwambazimana

, Midonzi

, Twagirumugabe

, Minega

, et al. (2005). A Rwandan altitude blood gas, acid-base and hemoglobin study. Clin Chim Acta, 357:86–87.

15.

, Babb

, Sivieri

, Resaland

, Karlsen

, Stray-Gundersen

, and Levine

. (2006). Urine acid–base compensation at simulated moderate altitude. High Alt Med Biol, 7:64–71.

16.

, Chen

, Wang

, Gen

, Yang

, Kubo

, et al. (1994). Higher exercise performance and lower VO2max in Tibetan than Han residents at 4,700 m altitude. J Appl Physiol, 77:684–691.

17.

Goldfarb-Rumyantzev

, and Alper

. (2014). Short-term responses of the kidney to high altitude in mountain climbers. Nephrol Dial Transplant, 29:497–506.

18.

González-García

, Barrero

, and Maldonado

. (2004). Exercise limitation in patients with chronic obstructive pulmonary disease at the altitude of Bogota (2640 m) breathing pattern and arterial gases at rest and peak exercise. Arch Bronconeumol, 40:54–61.

19.

Grocott

, Martin

, Levett

, McMorrow

, Windsor

, and Montgomery

. (2009). Arterial blood gases and oxygen content in climbers on Mount Everest. N Engl J Med, 360:140–149.

20.

Hackett

, and Roach

. (2012). High-altitude medicine. In: Wilderness Medicine. Auerbach

, ed. Mosby Elsevier, Philadelphia PA.pp. 1–2. Available at www-clinicalkey-es.pbidi.unam.mx:2443/#!/content/book/3-s2.0-B9781437716788000015 (accessed on February 9, 2016).

21.

Hansen

, Stelter

, and Vogel

. (1967). Arterial pyruvate, lactate, pH, and PCO2 during work at sea level and high altitude. J Appl Physiol, 23:523–530.

22.

Kumar

, and Karon

. (2008). Comparison of measured and calculated bicarbonate values. Clin Chem, 54:1586–1587.

23.

Kurtz

, Kraut

, Ornekian

, and Nguyen

. (2008). Acid-base analysis: a critique of the Stewart and bicarbonate-centered approaches. Am J Physiol Renal Physiol, 294:F1009–F1031.

24.

Lolekha

, Boonlert

, Kost

, Vanavanan

, and Lolekha

. (2003). Comparative study of values of calculated bicarbonate and measured total carbon dioxide content. Point Care, 2:135–143.

25.

Loniewski

, and Wesson

. (2014). Bicarbonate therapy for prevention of chronic kidney disease progression. Kidney Int, 85:529–535.

26.

Márquez-González

, Pámanes-González

, Márquez-Flores

, Gómez-Negrete

, Muñoz-Ramírez

, and Villa-Romero

. (2012). Lo que debe conocerse de la gasometría durante la guardia. Rev Mex Ins Seguro Soc, 50:389–396.

27.

Monge

, Lozano

, and Carcelen

. (1964). Renal excretion of bicarbonate in high altitude natives and in natives with chronic mountain sickness. J Clin Invest, 43:2303–2309.

28.

Muñoz-Arizpe

, Escobar

, and Medeiros

. (2012). Over-diagnosis of renal tubular acidosis in Mexico. Rev Invest Clin, 64:399–401.

29.

Nasir

, Sthaneshwar

, Yunus

, and Yap

. (2010). Comparing measured total carbon dioxide and calculated bicarbonate. Malays J Pathol, 32:21–26.

30.

Nyenwe

, and Kitabchi

. (2011). Evidence-based management of hyperglycemic emergencies in diabetes mellitus. Diabetes Res Clin Pract, 94:340–351.

31.

Paralikar

, and Paralikar

. (2010). High-altitude medicine. Indian J Occup Environ Med, 14:6–12.

32.

Pérez Padilla

, and Vázquez García

. (2000). Estimation of gasometric values at different altitudes above sea level in Mexico. Rev Invest Clin, 52:148–155.

33.

Purkayastha

, Ray

, Arora

, Chhabra

, Thakur

, Bandopadhyay

, et al. (1995). Acclimatization at high altitude in gradual and acute induction. J Appl Physiol, 79:487–492.

34.

Rico

, Gómez

, Meneses

, Múgica

, and Pérez

. (1998). Valores gasométricos en población geriátrica. Análisis en dos diferentes altitudes. Rev Inst Nal Enf Resp Mex, 11:25–28.

35.

Rico

, Urias

, Barquera

, Ochoa

, Padilla

, and Meneses

. (2001). Valores espirométricos y gasométricos en una población geriátrica sana, a diferentes alturas sobre el nivel del mar, en la República Mexicana* Estudio multicéntrico. Rev Inst Nal Enf Resp Mex, 14:90–98.

36.

Rose

, and Post

. (2001). Clinical Physiology of Acid-Base and Electrolyte Disorders. McGraw-Hill, New York City.

37.

Secretaría de Salud. (2010). Diagnóstico y Tratamiento del Desequilibrio AcidoBase. Available at www.cenetec.salud.gob.mx/interior/gpc.html (accessed in June 15, 2015).

38.

Theodore

. (2015). Arterial blood gases. In: UpToDate. Manaker

, ed. UpToDate, Waltham, MA, (accessed on June 15, 2015).

39.

United Nations Demographic Year Book. (2013). Estimates of population and its percentage distribution, by age and sex and sex ratio for all ages for the world, major areas and regions: 2013. UNDYB, 55–67. http://unstats.un.org/unsd/demographic/products/dyb/dybsets/2013.pdf (accessed in June 16, 2015).

40.

Wright

, Beazley

, Bradwell

, Chesner

, Clayton

, Forster

, et al. (2004). Medroxyprogesterone at high altitude. The effects on blood gases, cerebral regional oxygenation, and acute mountain sickness. Wilderness Environ Med, 15:25–31.

41.

Zubieta-Calleja

, Zubieta-Castillo

, Zubieta-Calleja

, Ardaya-Zubieta

, and Paulev

. (2011). Do over 200 million healthy altitude residents really suffer from chronic Acid-base disorders?. Indian J Clin Biochem, 26:62–65.