Calcium Carbonate for Elemental Lead Ingestions: Effect of Alkalinization on Elemental Lead Solubility in a Simulated Gastric Environment

Abstract

Introduction

Acute ingestion of elemental lead foreign bodies has resulted in multiple pediatric deaths. Elemental lead is relatively insoluble at alkaline pH. Furthermore, calcium decreases lead absorption by interfering with the lead absorptive receptor. We hypothesize that alkalinization of gastric fluid with an oral calcium-containing agent, such as calcium carbonate, will decrease lead solubility, thus reducing the potential for systemic lead absorption and toxicity.

Methods

This was an in vitro controlled study. One lead sphere (00 buckshot, cast 30 days prior) was randomly placed in each of ten tubes containing 20 mL simulated gastric fluid, with five tubes having 500 mg calcium carbonate added at 20 min and 140 min. We measured the fluid pH and the lead concentrations hourly for 4 h. We compared the median amount of total lead liberated after 4 h between the two groups using the Mann-Whitney U test.

Results

The pH of the gastric fluid only tubes remained 1 at every measurement, and the pH of the gastric fluid + calcium carbonate tubes was 6 at every measurement. At hour 4, the total amount of lead liberated in the soluble fraction in the control group vs the calcium carbonate group was 850 vs 12.4 mcg (95% CI for absolute difference: 605–964 mcg; p = 0.0079).

Conclusions

Calcium carbonate antacid alkalinizes gastric fluid pH and dramatically decreases the total amount of solubilized lead by 60-fold. This project lends foundational evidence to a low-cost, widely available, pre-hospital strategy to decrease lead absorption after acute elemental lead ingestions.

Electronic supplementary material

The online version of this article (10.1007/s13181-020-00811-6) contains supplementary material, which is available to authorized users.

Introduction

Lead remains one of the most common pediatric environmental poisonings in the USA [1]. Most lead research has focused on chronic ingestions and toxicity. However, significantly elevated blood lead levels and death have resulted from acute ingestion of foreign bodies containing elemental lead [2–6]. The rate of pediatric foreign body ingestions from 1995 to 2015 has increased over 90% [7]. Most recently, foreign body ingestions accounted for over 90,000 human exposures in 2018, with over 60,000 of these occurring in children under 5 years of age [8].

Recommendations for managing ingestions of elemental lead-containing foreign bodies are based on minimal evidence and at times seem contradictory to standard teachings. A 2001 abstract suggests a small amount of lead in water may be bound by activated charcoal [9]; this is the basis for POISINDEX® (a nationally used treatment database) to recommend treatment with activated charcoal [10]. This is unusual as charcoal is not believed to bind most metals well, with the exception of thallium (a heavy metal adjacent to lead on the periodic table), which has been shown to bind charcoal in small amounts [11, 12]. Moreover, activated charcoal may obscure visualization of the foreign body if endoscopy is performed. For ingestions of a small foreign body containing elemental lead, some resources recommend allowing it to pass while tolerating some absorption. However, children are a vulnerable group as they are thought to both absorb and retain more lead than adults, predisposing them to greater lead neurotoxicity [13]. Several case reports demonstrate that elevated blood levels and toxicity can result from acute elemental lead ingestion [2–6], suggesting that urgent removal of the object or whole bowel irrigation may be better options. While endoscopy or whole bowel irrigation can be performed to retrieve the object or enhance passage, respectively, these methods are not without risks, such as perforation and aspiration; additionally, whole bowel irrigation has never been shown to change passage time of lead objects. Finally, there is often a delayed time to endoscopic foreign body removal due to institutional capabilities and specialist availability.

Elemental lead is soluble in an acidic environment but relatively insoluble at alkaline pH. Alkalinizing the stomach environment may be beneficial since it is the most acidic portion of the gastrointestinal tract and thus where elemental lead would be expected to be the most soluble [14]. Soluble lead is absorbed in the duodenum [15], so lead absorption may be decreased if the lead foreign body only minimally solubilizes in the stomach. Calcium also decreases lead absorption by interfering with the lead absorptive receptor in the duodenum [15], so an oral calcium-containing alkalinizing agent such as calcium carbonate (CaCO₃; commonly available as TUMS®) may be an ideal agent for decreasing lead solubility as well as lead absorption. The purpose of this study is to characterize the kinetics of the solubility of elemental lead in gastric contents and to characterize alterations in lead solubility with the addition of an alkalinizing agent, calcium carbonate. We hypothesize that calcium carbonate will decrease lead solubility in a simulated gastric environment.

Materials and Methods

The lead foreign bodies were spheres cast by the investigators using 99.9% lead nuggets purchased from a metal supplier (RotoMetals, San Leandro, CA) using a 00 aluminum buckshot mold (Lee Precision, Hartford, WI). The spheres were stored for 30 days at room temperature to ensure that all had equal amounts of external oxidation. On the day of the experiment, 15 spheres that all visually appeared similar were measured with an analytical balance. The ten spheres closest in weight were chosen for the experiment. TUMS® Mint Regular 500 mg tablets were used as the source of CaCO₃. Each tablet was pulverized and weighed 1.2 g.

On the day of the experiment, we created a simulated gastric environment by mixing 1 L of simulated gastric fluid (0.2% (w/v) sodium chloride in 0.7% (v/v) hydrochloric acid) and 3.2 g of pepsin (Ricca Chemical Company®, Arlington, TX), which meets the US Pharmacopeia specifications for a simulated gastric fluid test solution. A sample of the gastric fluid was tested for lead, and the baseline pH was measured with pH paper (pH range 1 to 14).

Ten trace-metal free centrifuge tubes were prepared with 20.0 mL of gastric fluid, each in a warm water bath at 37 °C, placed on a shaker circling at 25 RPM. This RPM was chosen because a model analyzing fentanyl patch absorption used 24 RPM to simulate gastric motility [16]. At time zero, a randomly chosen lead sphere was introduced into each tube. At both 20 minutes and 2 hours 20 minutes from lead sphere placement, 500 mg of calcium carbonate was added to tubes 6–10, and the tubes were inverted seven times and returned to the shaker. Every hour, 2 mL of the liquid contents was collected in trace-metal free microcentrifuge tubes using trace-metal free pipette tips, and 0.1 mL of the contents was tested for pH; 2.1 mL of fresh gastric fluid was added to replace the volume removed. Figure 1 depicts our protocol and timeline.

Fig. 1.

Study protocol and timeline.

A volume of 20 mL of gastric fluid was chosen to approximate the higher range of a toddler’s (10–15 kg) gastric content volume. Toddlers are at an appropriate age to possess the motor skills to explore their environments and obtain potentially dangerous foreign objects; moreover, these objects will have the highest milligram per kilogram hazard to them, and lead toxicity has greater sequelae in this age group [13]. Studies demonstrate average gastric fluid volume in children ranges from 0 to 4 mL/kg. One study of 212 children had an average volume of 0.469 mL/kg (0–2.663 mL/kg) [17], while another study of 661 children showed the gastric fluid volume was on average 0.4 mL/kg, with the 95th percentile at 1.25 mL/kg and an upper limit of 4.1 mL/kg [18].

With the dearth of prior evidence, there was no expected lead solubility in simulated or actual gastric environments, so no power calculation could be performed. However, we determined a relevant amount of solubilized lead to be 2 mcg/mL (in 20 mL) based on the following estimation. The World Health Organization estimates the total blood volume of a 10-kg 15-month-old to be about 8 dL [19]. The Centers for Disease Control and Prevention recommends action at a blood lead level of 5 mcg/dL [1]. Assuming no other compartments, this correlates with a minimum of 40 mcg of lead needing to be absorbed to raise the blood lead level to 5 mcg/dL. The soluble lead content of the samples were analyzed at NMS Laboratories (Willow Grove, PA) using inductively coupled plasma spectroscopy (1 ng/mL lead detection limit) [20].

We evaluated the lead concentrations over time for the two groups (control and calcium carbonate). Our primary outcome is the total amount of lead at hour 4 in the soluble fraction. Our null hypothesis is that the difference between the total amount of lead solubilized at hour 4 in both groups will be equal to zero. For the expected non-parametric data, we calculated median and interquartile range (IQR) for the concentrations at each time point as well as the total amount liberated at hour 4, and we compared the primary outcome for these two groups using a Mann-Whitney U test with significance defined as a two-tailed p < 0.05.

Because 0.021 dL were removed during each time point, the total amount of lead solubilized was equal to all that was removed plus the amount in the tube at 4 h. Thus, the total amount liberated at the end of the protocol was computed as [Lead concentration @ Hour 4]*0.2 + [Lead concentration @ Hour 3]* 0.021 + [Lead concentration @ Hour 2]* 0.021 + [Lead concentration @ Hour 1]* 0.021, with lead concentrations measured in mcg/dL and expressed as median and IQR. All statistics were performed using Microsoft Excel® (Redmond, WA).

Results

The lead spheres all weighed 3.3 g and had a diameter of 8.1 mm with a surface area of 2.06 cm². The pH of the gastric fluid was 1, and the lead concentration was < 1 ng/mL (undetectable).

The pH of the control tubes remained 1 at each hourly measurement. The pH of the calcium carbonate tubes was 6 at each hourly measurement. There was a rapid rise in the lead concentrations in the control tubes compared with the calcium carbonate tubes (see Fig. 2 and Table 1). Every tube’s lead concentrations are given in Online Resource 1. The median total amount of lead liberated in the soluble fraction in the control group vs the calcium carbonate group at hour 4 was 850 mcg (IQR 802–902 mcg) vs 12.4 mcg (IQR 10.4–12.5 mcg), with 95% CI for absolute difference of 605–964 mcg, p = 0.0079. As only the surface of the object was involved in lead solubility, it is estimated that about 413 mcg of lead was solubilized per cm² of object surface area by the end of 4 hours.

Fig. 2.

Median lead concentration (IQR) (mcg/dL) vs time (hour). Arrows indicate 500 mg CaCo3 administration, each point n = 5

Table 1.

Median lead concentration (mcg/dL), n = 5 controls and CaCO₃

	Hour 1​ (IQR)	Hour 2​ (IQR)	Hour 3​ (IQR)	Hour 4​ (IQR)
CaCO3​	140​ (120–260)	130​ (110–140)	40​ (32–65)	18​ (18–35)
Control​	1900​ (1900–2000)	2700​ (2500–2900)	3300​ (3100–3400)	3400​ (3200–3700)

Because there was a decline in the lead concentrations in the calcium carbonate tubes (suggesting precipitation of lead out of solution as opposed to merely halting further solubility), a post hoc analysis using linear extrapolation from the hour 1 concentration in the control tube was performed to estimate the amount of lead solubilized in the first 20 minutes as a worst-case scenario of maximally solubilized and potentially absorbed lead prior to the addition of the first calcium carbonate dose. This suggested that a median of approximately 127 mcg (IQR 127–133 mcg) of lead, or about 61 mcg per cm² of surface area, was solubilized in the first 20 min. Compared with the final amount at hour 4, there is still a difference of 850–127 = 723 mcg, with 95% CI for absolute difference of 491–850 mcg, p = 0.0079.

Discussion

Elemental lead exposed to the atmosphere develops a dull coating consisting of lead oxide (PbO) and lead basic carbonate (PbCO₃·Pb(OH)₂) from a reaction between the lead metal, oxygen, carbon dioxide, and water, which limits further oxidation. The chemical properties of these oxidative surface products suggest they are the most likely source of the soluble lead observed in this experiment and not the elemental metal. Elemental lead oxidizes with exposure to hydrochloric acid, producing Pb²⁺ (and hydrogen gas), but this reaction is relatively slow at gastric pH and probably not relevant in this experimental setup. The much faster non-oxidative/reduction reaction(s) are the markedly increased solubilization of lead oxide and lead basic carbonate that occur in an acidic environment presented below (with simplification).

$$\mathrm{PbO}+{\text{2H}}^{+}\to {\mathrm{Pb}}^{2+}+{\mathrm{H}}_2\mathrm{O}$$

$${\text{PbCO}}_3·\mathrm{Pb}{\left(\mathrm{OH}\right)}_2+{\text{4H}}^{+}\to {\text{2Pb}}^{2+}+{\mathrm{H}}_2{\mathrm{CO}}_3+{\text{2H}}_2\mathrm{O}$$

Another component to the solubility of lead in gastric fluid is the formation of the soluble complex tetrachloroplumbate(II) ion (PbCl₄²⁻), which is responsible for a counterintuitive increase in lead solubility with increasing chloride concentration. This is because PbCl₂ has limited aqueous solubility and an increase in chloride would be expected to decrease the amount of lead in solution if there was no formation of this complex ion.

$${\mathrm{Pb}}^{2+}+2{\mathrm{Cl}}^{-}\leftrightarrow {\text{PbCl}}_2$$

$${\text{PbCl}}_2+2{\mathrm{Cl}}^{-}\leftrightarrow {{\text{PbCl}}_4}^{2-}$$

The maximal absolute release of lead appears to occur in the first hour, with approximately 45% (380 mcg) of the total lead solubilized in the control group being liberated in the first hour. Lead continued to solubilize further, but it appeared to plateau by hour 4. This initial rise and subsequent plateau suggest a rapid solubilization of the outer oxidized coating, possibly followed by the much slower oxidation of the lead metal. This matches the known physical chemistry properties of metallic lead and its relevant salts in a hydrochloric acid environment. This would mean a dependence on the surface area and degree of surface oxidation as to how many lead ions are initially released rather than the absolute weight of the lead object. Alternatively, this plateau may be a function of our non-continuous gastric secretion model and saturation of the gastric fluid.

The declining lead concentrations in the calcium carbonate arm suggest that alkalinization will precipitate lead out of solution. Thus, alkalinization is time-sensitive, and calcium carbonate therapy would therefore be best performed pre-hospital. Of note, after just one dose of CaCO₃, there was about × 8000 more calcium present in the calcium carbonate tubes to compete for lead absorption compared with the maximal calculated amount of lead at 20 min which mostly precipitated out of solution.

We chose 500 mg of calcium carbonate because this is the dose in a single TUMS® Regular tablet that is expected to be most widely available to the public. In a randomized control trial in healthy adults, 1360 mg calcium carbonate and 160 mg magnesium carbonate used in one arm achieved a pH > 3 in a median of 5.8 minutes [21]. However, the alkalinization effect’s duration was less than half that of a H2-receptor antagonist, likely due to continued acid secretion. Therefore, we gave a second dose 2 hours after the first dose. Since lead foreign bodies are more often pediatric ingestions, we started with a smaller dose than the aforementioned adult study. The total dose of 1000 mg calcium carbonate is not expected to cause toxicity as its 400 mg of elemental calcium is below the 1000 mg upper level of intake for the youngest children and well below the 2500 mg upper level of intake for a child aged 1 to 8 years of age, as set by the Institute of Medicine [22].

This study has several limitations. One is that this is an in vitro model using simulated gastric fluid. Cells and secreted fluid in the gastrointestinal tract may have compounds that affect absorption beyond solubility. We also did not test lead absorption, only solubility, though it is logical that if lead does not solubilize, it will not be absorbed. Our hourly samples and repletions crudely approximated gastric secretion. Finally, our calcium carbonate arm demonstrated a fall in the soluble lead fraction, which more likely than not indicated a precipitation of the lead ions, probably as the highly insoluble lead carbonate. The potential absorption from the GI tract of the precipitate is not known, though our post hoc analysis shows that even if this precipitate was to be absorbed, there remains a significant difference between the two groups.

Conclusion

In conclusion, we have shown that alkalinization of simulated gastric contents decreased lead solubility, thereby providing foundational evidence of a low-cost, widely available, time-sensitive, pre-hospital strategy to potentially decrease lead absorption after elemental lead ingestions. Further research using an in vivo model or animal study would shed additional light on the matter. The impact of gastric alkalinization on other interventions, such as visualization during endoscopy, should be considered as well. Based on the current literature as well as the efficacy, safety, and cost of this intervention, we suggest that calcium carbonate antacids be strongly considered early in suspected acute lead foreign body ingestions, especially if there is an anticipated delay in removal of the object.

Funding

This study was supported by a charitable foundation, the Atrium Health Foundation Dr. John A. Marx Fund.

Compliance with Ethical Standards

Conflict of Interest

The authors declare that they have no conflict of interest.