Acidosis is a frequent problem in critically-ill neonates. Buffers, such as sodium bicarbonate, are often used to treat metabolic acidosis. However, evidence to support the use or efficacy of this therapy is lacking.

The etiology of a low pH must be understood to treat infants appropriately. Respiratory acidosis (increased CO₂ on a blood gas and normal or near normal serum bicarbonate concentration) can only be treated by improving ventilation. Buffers will not help in this case, and may make the situation worse because infusion of NaHCO₃ results in the immediate formation of CO₂. For every mole of proton neutralized by bicarbonate, an equimolar amount of CO₂ is produced.

H⁺ + HCO₃^- «   H₂CO₃ «  H₂O + CO₂

The futility of using NaHCO₃ in a situation where ventilation is inadequate can be appreciated by the Henderson-Hesselbalch equation:

pH - pK₁ + log [HCO₃^-]/[CO₂]  (pK₁ = 6.1)

When ventilation is impaired, use of NaHCO₃ will move the pH toward the pK of the equation, which is 6.1. In order to achieve a pH of 7.4, the molar ratio of HCO₃ to CO₂ must be 20:1. Thus, correction of acidosis depends on the removal of CO₂. CO₂ elimination depends on minute ventilation and pulmonary blood flow. Infusion of NaHCO₃ in a patient with inadequate minute ventilation will worsen acidosis from CO₂ accumulation and shift of the Henderson-Hesselbalch equation to the left. Compared to bicarbonate, CO₂ rapidly crosses cell membranes, leading to intracellular acidosis. The bicarbonate lags behind in the vascular space, causing an increase in arterial pH. The intracellular acidosis associated with the use of NaHCO₃ may not be reflected in the arterial blood gases we follow so carefully in our patients. 

The etiology of a metabolic acidosis (low pH associated with low serum bicarbonate and negative base deficit, but normal or near normal pCO2) must be considered to determine the need to treat and appropriate intervention. In all cases, giving buffer will at best only partially and transiently improve the arterial pH; the underlying problem generating the acidosis must be addressed.

Metabolic acidosis is the consequence of one of three fundamental mechanisms:

1.      Loss of base - via renal or gastrointestinal route

2.      Intake of more acid then the kidneys can excrete (e.g. high protein diet or renal insufficiency)

3.      Abnormal metabolism resulting in increased endogenous acid:

a.       Inorganic acids (e.g. nitrates, sulfates, phosphates) from rapid tissue catabolism in the very ill

b.      Organic acids due to incomplete oxidation of fuels (e.g. lactate, acetoacetate, methylmalonate)

Loss of Base: This can be the result of renal tubular acidosis or chronic diarrhea. In both cases, hyperchloremia is common, hypokalemia may occur, and the serum anion gap, defined as [Na] – ([Cl] + [CO₂]), is not increase above normal (6-15 mmol/L). 

Conditions associated with increases in the anion gap due to abnormal concentrations of anions (inorganic or organic) include:

Excess intake of acid: Late metabolic acidosis of prematurity is an example of an increase in [H⁺] due to intake in excess of renal clearance. In this situation, the ability of immature kidneys to excrete an acid load associated with a net-acid diet is exceeded. The urinary pH usually falls below 5.5 and growth is impaired. This was more common when premature infant formulas were casein based, but it still occurs. The anion gap is increased because of accumulating inorganic anions.

Abnormal metabolism: The rapidly catabolic state of an extremely ill neonate is an example of a condition that increases serum inorganic acids (phosphates, nitrates, etc.) which contribute to the metabolic acidosis seen in such infants.

The most common cause of metabolic acidosis is increased organic acid, usually lactic acid, from abnormal metabolism. Lactic acid is increased in the circulation when mitochondria are unable to convert lactate to carbon dioxide, water, and ATP. The mechanisms include (a) mitochondria that are not functioning properly (e.g., Leigh’s disease, cyanide poisoning) or mitochondrial that are deprived of oxygen. Common clinical conditions associated with oxygen deprivation at the tissue or cellular levels are hypoxemia, compromised cardiac output (cardiogenic, septic, hypovolemic shock), severe anemia, or abnormal hemoglobin (hemoglobinopathy, excess methemoglobin (as can be seen with nitric oxide administration), or carbon monoxide poisoning). 

These widely differing underlying causes of metabolic acidosis require different approaches to treatment. If the problem is chronic loss of base, then administration of base (such as acetate in the hyperalimentation solution, sodium citrate or dilute sodium bicarbonate) may be appropriate. If the problem is poor perfusion, addition of base will do little to correct the problem. Indeed in the face of poor perfusion, administration of base is likely to cause intracellular acidosis and venous hypercarbia. Successful treatment of metabolic acidosis, therefore, is highly dependent on the proper identification and specific treatment of the underlying process. Once that is known, a logical and effective approach to therapy becomes possible.

The etiology of the acidosis also determines the threshold at which you would consider administering buffers to a patient. Buffers are administered to patients with congenital heart disease because acidosis impairs cardiac performance at a pH < 7.2. However, data that administration of bicarbonate even in this circumstance actually improves outcomes are lacking. Indeed, there is evidence that bicarbonate administration may improve the arterial pH while simultaneously lowering intramyocardial pH and worsen myocardial performance. Much depends on the respiratory status of the patient. In general, bicarbonate should only be given in situations when adequate ventilation and CO2 elimination are assured.

In extremely premature infants with hyperchloremia, the risk-benefit ratio differs. Hyperchloremic acidosis is better tolerated and pharmacologic buffers are hypersosmolar and associated with an increased risk of IVH. Therefore, buffers are reserved for a pH of < 7.2 and a base deficit of > -8 to -10. Even in those circumstances, bicarbonate may not be the drug of choice because of its hyperosmolarity. Once again, adequate ventilation remains essential in order for bicarbonate to act as a physiological buffer. A 2005 Cochrane Systematic Review¹⁶ attempted to evaluate the available evidence from RCT of preterm infants with metabolic acidosis treated with infusion of base versus placebo. Only one small RCT was found which demonstrated that addition of NaHCO₃ to the IV infusion of an acidemic infant was no more effective in decreasing morbidity (intracranial hemorrhage) and mortality than glucose and water alone. Furthermore, the pH was corrected just as quickly without the NaHCO₃ as with it. Slow correction of acidosis by replacing chloride with acetate in the IV fluids may be a safer choice, although the data for this therapy are likewise lacking.

Also recall that a low serum concentration of bicarbonate alone does not mean that the infant is deficient in bicarbonate. Bicarbonate concentration falls naturally when [H+] rises, since carbonic acid is formed and is converted to carbon dioxide and water. This happens when an excess of another weak acid such as lactic acid is present in excess. The usual physiologic response to metabolic acidosis is an attempt to compensate by increasing ventilation and reducing pCO₂, returning pH toward normal.  When the pCO₂ is not reduced as much as expected, a mixed acidosis is present and more than one underlying mechanism leading to acidosis should be considered.

Bicarbonate

If NaHCO₃ is to be given at all, it should be administered as a dilute solution (½ strength bicarbonate 0.5 meq NaHCO₃^- per mL) only after adequate oxygenation and ventilation have been ensured as bicarbonate generates CO2.

If you wish to correct a known base deficit:

# meq HCO₃^- = 0.3 x baby’s weight in kg x base deficit

Bicarbonate is hypersomolar and should be given slowly via reliable iv access. In a non-emergent situation, give over >1 hour. Rapid administration has been associated with IVH in small premature infants. Possible adverse effects include fluctuations in cerebral blood flow, diminished oxygen delivery to tissues and worsening intracellular acidosis, in addition to IVH. Administering bicarbonate also gives a significant sodium load.

For patients with a very large base deficit and a mixed respiratory/metabolic acidosis, first correct the ventilatory problem as this may be accompanied by improvement in the metabolic acidosis as well. If base is deemed appropriate, consider giving ½ the above dose and checking a repeat blood gas to avoid overshooting the target pH and creating an iatrogenic metabolic alkalosis. Lastly, it must be acknowledged that current recommendations for dose, dilution and rate of administration are largely arbitrary.

NOTE: In the delivery room, bicarbonate is no longer recommended by the NRP as part of routine neonatal resuscitation.

THAM (Tromethamine acetate)

Data on the use of THAM in the neonate are entirely lacking. Theoretical advantages include that THAM doesn’t increase CO₂ and it doesn’t increase the sodium load. However, THAM is extremely hyperosmolar, can cause hypoglycemia, and requires renal excretion and will accumulate in patients with renal insufficiency. THAM is occasionally used in patients with a mixed respiratory and metabolic acidosis in whom ventilation cannot be easily improved and in patients with combined hypernatremia and metabolic acidosis.

To administer THAM, give:

# mL THAM = 1.1 x baby’s weight in kg x base deficit

However, this may amount to a very large volume; alternately

# mL THAM = 3.3 x baby’s weight

may be administered and the patient’s blood gas rechecked.

NEVER give THAM through a UVC. It can cause hepatic necrosis. UAC administration should also be avoided.

Give THAM slowly (over at least 30 minutes) through a PICC line, BROVIAC® catheter, or PIV. Monitor the infusion site closely if using a PIV.

References:

Beveridege, CJ, Wilkinson AR. Sodium bicarbonate infusion during resuscitation of infants at birth. Cochrane Database Syst Rev (1):CD004864, 2006.

Corbet AJ, Adams LM, Kenny JD et al. Controlled trial of bicarbonate therapy in high-risk premature newborn infants. J Pediatr 91:771-6, 1977.

Dixon H, Hawdins, K Stephenson T. Comparison of albumin versus bicarbonate treatment for neonatal metabolic acidosis. Eur J Pediatr 158:414-5, 1999.

Hein HA. The use of sodium bicarbonate in neonatal resuscitation: Help or Harm? Pediatrics 91:496-7, 1993.

Lawn CJ, Weir FJ McGuire W. Base administration or fluid bolus for preventing morbidity and mortality in preterm infants with metabolic acidosis. Cochrane Database Syst Rev. (2):CD003215, 2005

Lokesh L, Kumar P, Murski S, Narang A..  A randomized controlled trial of sodium bicarbonate in neonatal resuscitation-effect on immediate outcome. Resuscitation 60:219-23, 2004.

Niermeyer S, Kattwinkel J, Van Reempts P et al. International Guidelines for Neonatal Resuscitation: An Excerpt From the Guidelines 2000 for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care: International Consensus on Science. Pediatrics 106 e29, 2000.

Van Alfen-Van der Velden AA, Hopman JC, Klaessens, JH et al. Effects of Rapid versus Slow Infusion of Sodium Bicarbonate on Cerebral Hemodynamics and Oxygenation in Preterm Infants. Biol Neonate 90:122-7, 2006

Young TE, Mangrum OB. Neofax 2001: A Manual of Drugs Used in Neonatal Care, 14^th ed. Acorn Publishing. pgs. 184-6.

BROVIAC® is a registered trademark of C.R.Bard, Inc. and its related company, BCR, Inc.