Prevalence and clinical implications of heightened plastic chemical exposure in pediatric patients undergoing cardiopulmonary bypass

Abstract

Background:

Phthalate chemicals are used to manufacture plastic medical products, including many components of cardiopulmonary bypass (CPB) circuits. We aimed to quantify iatrogenic phthalate exposure in pediatric patients undergoing cardiac surgery and examine the link between phthalate exposure and postoperative outcomes.

Study Design and Methods:

The study included pediatric patients undergoing (n=122) unique cardiac surgeries at Children’s National Hospital. For each patient, a single plasma sample was collected preoperatively and two additional samples were collected postoperatively upon return from the operating room and the morning after surgery. Concentrations of di(2-ethylhexyl) phthalate (DEHP) and its metabolites were quantified using ultra high-pressure liquid chromatography coupled to mass spectrometry.

Results:

Patients were subdivided into three groups, according to surgical procedure: (1) cardiac surgery not requiring CPB support, (2) cardiac surgery requiring CPB with a crystalloid prime, and (3) cardiac surgery requiring CPB with red blood cells (RBCs) to prime the circuit. Phthalate metabolites were detected in all patients, and postoperative phthalate levels were highest in patients undergoing CPB with an RBC-based prime. Age-matched (<1 year) CPB patients with elevated phthalate exposure were more likely to experience postoperative complications. RBC washing was an effective strategy to reduce phthalate levels in CPB prime.

Discussion:

Pediatric cardiac surgery patients are exposed to phthalate chemicals from plastic medical products, and the degree of exposure increases in the context of CPB with an RBC-based prime. Additional studies are warranted to measure the direct effect of phthalates on patient health outcomes and investigate mitigation strategies to reduce exposure.

1 |. INTRODUCTION

Phthalates are chemical additives used to manufacture consumer and medical-grade plastics.^1,2 Di(2-ethylhexyl) phthalate (DEHP) is one of the most commonly used plasticizers in polyvinyl chloride (PVC) plastics; it functions by embedding between PVC polymers to increase spacing and impart flexibility. DEHP often contributes 40%–80% of the finished weight of medical-grade tubing and blood storage bags.^3,4 Notably, phthalates are not covalently bound to the PVC matrix, which makes them highly susceptible to leaching—particularly in the presence of lipophilic solutions.^5–7 Historically, phthalate leaching was considered beneficial to red blood cell (RBC) storage, as plasticizers interact directly with RBC membranes thus imparting resistance to storage hemolysis^8–11 however safety concerns surrounding phthalate exposure have increased since the 1970’s.^4,12–14

Biomonitoring studies indicate that the general population incurs daily phthalate exposures, with DEHP metabolites detected in nearly 80% of the United States population.^15,16 Furthermore, a recent study in infants with congenital heart defects reported a correlation between environmental phthalate exposure and developmental delays in language and motor skills.¹⁷ Heightened phthalate exposure from medical equipment is even more concerning, particularly for neonatal and pediatric intensive care unit patients (NICU, PICU).^18–22 Multiple studies have documented increased phthalate exposures in NICU patients^21–25 and phthalates have been found to localize to the hearts of infants receiving umbilical catheterization and/or blood transfusions.^26,27 The latter is concerning since low-dose environmental phthalate exposure is associated with an increased risk of all-cause and cardiovascular mortality.²⁸ Moreover, experimental studies report a direct causation between elevated phthalate exposure and cardiotoxicity.^29–32

The primary goal of this study was to quantify phthalate exposure, including DEHP and its metabolites, in pediatric patients undergoing cardiac surgery with/without cardiopulmonary bypass (CPB). We anticipated that children undergoing CPB would be exposed to significant amounts of DEHP from blood products, CPB circuit components, and endotracheal tubes (Figure 1). We also aimed to investigate the timing of phthalate clearance in the postoperative period. Finally, we assessed possible correlations between higher phthalate exposures and postoperative complications.

FIGURE 1.

Phthalate exposure during pediatric cardiopulmonary bypass. Pediatric cardiac surgery patients undergoing cardiopulmonary bypass are exposed to phthalate plasticizers through contact with plastic medical devices manufactured with DEHP (indicated with blue outline), including blood product and crystalloid solution storage bags, tubing, cannulas, and reservoirs. The leached DEHP chemical (indicated with blue circles) can localize to the heart, where it can contribute to postoperative complications. DEHP, di(2-ethylhexyl) phthalate; ET, endotracheal tube; FFP, fresh frozen plasma; RBCs, red blood cells. [Color figure can be viewed at wileyonlinelibrary.com]

2 |. STUDY DESIGN AND METHODS

2.1 |. Study population

Experiments were conducted in accordance with human research guidelines; this study was approved by the Children’s National Hospital Institutional Review Board (Pro00009620). For children old enough to comprehend the study, the patient’s assent was obtained in addition to parental permission. The study population was divided into three groups (Figure 2). Group 1 “No CPB” served as a surgical control and included patients undergoing a thoracotomy without CPB (n = 11). These surgeries mostly included repairing coarctation of the aorta or pulmonary artery banding, wherein the only identifiable source of phthalate exposure was the endotracheal tube (syringes, foley catheters, central/arterial/peripheral lines regularly used for cardiac surgery at our institution are made from DEHP-free plastics). CPB patients were subdivided based on the CPB prime composition, since DEHP leaching is prevalent in RBC units but limited in crystalloid solutions devoid of lipids.¹⁰ Group 2 “Crystalloid prime” included CPB patients (n = 21) with a crystalloid based prime, with (n = 9) or without (n = 12) fresh frozen plasma (FFP) supplementation. As our institution uses a mix of DEHP-free and DEHP-plasticized FFP storage bags, we included subgroups in Group 2 to assess the impact of FFP on phthalate levels. Group 3 “RBC prime” included CPB patients with an RBC-based prime (n = 90). Patients received either unwashed RBCs (n = 33) or RBCs washed in a Sorin XTRA cell saver (n = 57).

FIGURE 2.

Overview of surgical groups. Pediatric cardiac surgery patients (n = 110) undergoing n = 122 unique operations were enrolled in this study. Patients were subdivided into those undergoing cardiac surgery without cardiopulmonary bypass (CPB) (Group 1, n = 11) or with CPB (Group 2, 3). CPB patients were further subdivided based on their CPB prime components. Group 2 (n = 21) included patients receiving a crystalloid prime with (n = 9) or without (n = 12) the addition of FFP while Group 3 (n = 90) included patients receiving RBC-based priming fluids. In the latter, patients received unwashed RBCs (n = 33) or washed RBCs (n = 57). [Color figure can be viewed at wileyonlinelibrary.com]

2.2 |. Cardiopulmonary bypass

At Children’s National Hospital, cardiac surgery prioritizes the use of third-generation leukoreduced, irradiated RBC units (both apheresis and whole-blood derived) that are stored for less than 8 days. For patients weighing <5 kg, RBCs were washed to remove extracellular potassium and lactate which accumulate during storage and following irradiation.^33,34 For patients weighing 5–15 kg, unwashed RBCs were used since it is assumed that lactate is more easily cleared in these patients. For patients weighing >15 kg, RBCs were not used in the prime, as long as the estimated postoperative hematocrit was deemed acceptable by the perfusionist. However, RBCs would still be used if significant hemodilution was expected and transfused if significant bleeding occurred. In a subset of patients, methylprednisolone was administered to reduce CPB-induced inflammation, which can be more significant in younger, smaller patients.³⁵ Neonates received methylprednisolone at approximately 2 a.m. on the morning of their surgery and 10 mg/kg methylprednisolone was added to their CPB prime.

2.3 |. Patient and RBC unit sample collection

Plasma samples were obtained from specimens collected for routine clinical tests. Efforts were made to minimize environmental phthalate contamination in the research laboratory, through the use of PVC- and DEHP-free tubes, tubing, and pipette tips. A preoperative (preop) sample was predominately collected the day before or morning of surgery. Postoperative samples were drawn upon arrival to the cardiac intensive care unit (postoperative day 0, “POD0”) and approximately 4 am on the morning after surgery (postoperative day 1, “POD1”). To evaluate whether washing RBC units altered phthalate levels, an in vitro study was performed (n = 4) wherein samples were collected before and after cell washing with a cell saver. Aliquoted samples were shipped on dry ice to the University of Colorado School of Medicine Metabolomics Core for quantification of DEHP and its metabolites (MEHP: mono(2-ethylhexyl)phthalate, MEHHP: mono(2-ethyl-5-hydroxyhexyl)phthalate, MECPP: mono(2-ethyl-5-carboxypentyl)phthalate), as previously described^36–38 and detailed in the Supplemental Methods.

Phthalate chemical concentrations were measured in both patient plasma and in samples collected from unwashed and washed RBC units. Plasma samples (as opposed to urine samples) were used due to clinical availability. Both our laboratory and other groups^38–43 have previously quantified phthalate concentrations in blood samples, which are a relevant biological matrix when evaluating exposure in the context of transfusion, CPB, and/or extracorporeal membrane oxygenation (ECMO). Both the parent chemical (DEHP) and its primary metabolites were quantified, as invasive clinical procedures can immediately result in high circulating phthalate levels. Since DEHP has a half-life of ~4–5 h in humans,^44–46 its presence in blood samples is sufficiently high enough for quantitation.^47–52 Accordingly, we reported the phthalate equivalent level (DEHP + MEHP + MEHHP + MECPP).⁵³ This approach aligns with recent efforts to quantify total environmental chemical exposure, as opposed to investigating each metabolite in isolation.⁵⁴

2.4 |. Postoperative outcome analyses

Clinical data was retrospectively collected from patients’ electronic medical records (Supplemental Methods). To limit age and surgical procedure as confounding factors, only patients <1 year old undergoing CPB were included in postoperative complication and intervention analysis (Group 2: n = 1, Group 3: n = 80).

2.5 |. Statistical analysis

Normality was assessed using a Shapiro–Wilk test and variance was assessed using an F-test (two groups) or Barlett’s test (three groups). The majority of the analyzed measurements had normal distribution; therefore, the results are reported as the mean ± standard deviation. For repeated measurements, normally distributed data were analyzed via one-way ANOVA with/without Geisser–Greenhouse correction (unequal variance). Nonparametric data were analyzed via a Friedman test. For independent measurements, normally distributed data were analyzed via one-way ANOVA with/without Welch’s correction (unequal variance). Nonparametric data were analyzed via Kruskal–Wallis ANOVA. A false discovery rate (FDR, 0.1 cutoff) was used to correct for multiple comparisons.^55–57 A two-tailed Student’s t-test was used to compare phthalate levels and postoperative complications with/without Welch’s correction (unequal variance). Spearman correlations were used to measure the association between independent surgical variables and equivalent levels. Multivariable analysis was implemented to adjust for covariates (e.g., age, race, sex, body surface area [BSA], steroid use, inpatient status, index flow rate, index prime volume, coolest venous temperature, CPB duration, cross-clamp duration, and time between the end of the procedure and sample collection). A p-value <.05 was considered statistically significant.

3 |. RESULTS

3.1 |. Patient demographics

Our study cohort included n = 110 individual pediatric patients (1 day–17 years old) undergoing n = 122 unique operations (Table 1). The timing of sample collection between patient groups was comparable as the median time from the end of surgery to POD0 sample collection was similar between patients (Group 1: 60 min, Group 2: 68 min, Group 3: 62 min). Median time from the end of surgery to POD1 sample collection was also comparable, ranging from 12.97 to 14.75 h. Time from the end of CPB to sample collection on POD0 (Group 2: 2.85 h, Group 3: 2.80 h) and POD1 (Group 2: 15.08, Group 3: 14.79) was also similar between crystalloid and RBC-based prime cohorts.

TABLE 1.

Patient demographics according to surgical group.

	No CPB	Crystalloid prime ± FFP	RBC prime
No. of cases	11	21	90
Age	8 days [2–233 days]	5.34 years [154 days–17.34 years]	93.5 days [1 day–9.26 years]
Weight (kg)	3.11 [2.08–6.11]	19.40 [7.60–81.80]	4.38 [2.15–29.40]
BSA (m2)	0.20 [0.14–0.32]	0.79 [0.35–1.91]	0.25 [0.16–0.32]
Inpatient (yes)	10 (90.9%)	1 (4.8%)	54 (60%)
Inpatient Preop LOS (days)	7[2–212]	1	9 [1–179]
CPB stop to POD0 (hours)	—	2.85 [1.68–4.68]	2.80 [1.60–6.47]
CPB stop to POD1 (hours)	—	15.08 [12.00–17.88]	14.79 [5.08–18.2]
Surgery stop to POD0 (min)	60 [26–88]	68 [21–126]	62 [32–284]
Surgery stop to POD1 (hours)	14.75 [9.37–17.80]	13.2 [9.73–16.43]	12.97 [5.85–16.75]
Time to transfer out of CICU (days)	28.98 [1.13–192.8]	2.01 [0.84–13.97]	5.07 [0.82–225.7]
Postop LOS (days)	44.57 [2.95–192.8]	4.75 [2.83–177.0]	9.97 [2.94–235.6]
Prime volume (mL)	—	405.8 [195.5–790.0]	113.6 [50.6–342.2]
Indexed prime volume (mL/m2)	—	453.1 [283.3–878.0]	951.4 [425.2–3435]
RBC volume in CPB prime (mL)	—	—	100 [50–155]
Indexed RBC volume in CPB prime (mL/m2)	—	—	400.0 [107.1–833.3]
Received RBCs during surgery (yes)	2 (18.2%)	4 (19.0%)	90 (100%)
Total RBC volume received during surgery (mL)	45 [40–50]	120 [70–846]	195 [80–440]
Indexed total RBC volume received during surgery (mL/m2)	216.0 [137.9–294.1]	325.0 [85.4–457.3]	866.4 [155.0–2200.0]
FFP in CPB prime (yes)	—	9 (42.9%)	90 (100%)
FFP volume in CPB prime (mL)	—	100 [75–100]	105 [45–150]
Indexed FFP volume in CPB prime (mL/m2)	—	127 [95.2–285.7]	400 [133.9–750.0]
Flow rate (L/min)	—	1.89 (0.85–4.58)	0.59 (0.38–2.41)
Indexed flow rate (L/min/m2)	—	2.40 [2.38–2.43]	2.41 [2.33–2.48]
10 mg/kg methylprednisolone (yes)	0 (0%)	0 (0%)	24 (26.7%)
CPB duration (min)	—	84.0 [42.0–257.1]	113.6 [50.6–342.2]
Cross-clamp (yes)	—	14 (66.7%)	83 (92.2%)
Cross-clamp duration (min)	—	41.7 [15.8–82.2]	70.1 [20.5–195.6]
Lowest venous temperature (°C)	—	29.4 [25.2–32.1]	26.8 [18.0–34.4]
Fluid balance (mL)	—	−270 [−942–271]	−78 [−778–337]
Inotrope score	0 [0–15]	0 [0–8]	0 [0–15]
Vasoactive inotropic score	1.425 [0–22.25]	0 [0–11]	5 [0–22]

Note: Values are reported as median [range] or n-value (percent). Inotrope score and vasoactive inotropic score were calculated at midnight the day of surgery. Abbreviations: BSA, body surface area; CICU, cardiac intensive care unit; CPB, cardiopulmonary bypass; FFP, fresh frozen plasma; LOS, length of stay; POD0, postoperative day 0; POD1, postoperative day 1; Postop, postoperative; Preop, preoperative; RBC, red blood cell (both washed and unwashed units).

3.2 |. Change in phthalate levels in cardiac surgery patients

Previous studies have reported routine phthalate exposure in the general population.^15,16 In agreement, we detected DEHP and its metabolites in all preoperative blood samples, regardless of inpatient or outpatient status (Figure 3, Supplemental Table 1). The preoperative equivalent level (sum of DEHP and its metabolites) was comparable across groups (No CPB: 1.37 ± 0.73 μM, Crystalloid prime: 1.42 ± 0.65 μM, Crystalloid prime + FFP: 1.59 ± 1.01 μM, RBC-based prime: 1.96 ± 2.56 μM, p > .05). The “No CPB” group had a minimal 0.8% increase in equivalent level after surgery (POD0), compared to the preoperative level (Figure 3e). Patients receiving crystalloid prime, crystalloid prime + FFP, or RBC-based prime had a 94%, 176%, or 481% increase in their equivalent level after surgery, respectively, compared to their preoperative levels (Figure 3j, o, t). Patients receiving an RBC-based prime had an increase in every phthalate measured: 193% increase in DEHP (Preop: 1.29 ± 1.51 μM, POD0: 3.79 ± 2.24 μM), 1077% increase in MEHP (Preop: 0.44 ± 1.32 μM, POD0: 5.12 ± 4.37 μM), 1203% increase in MECPP (Preop: 0.14 ± 0.28 μM, POD0: 1.85 ± 1.46 μM), and 592% increase in MEHHP (Preop: 0.09 ± 0.25 μM, POD0: 0.64 ± 0.43 μM; Figure 3p–s).

FIGURE 3.

Phthalate chemical concentrations increase in patient blood samples after cardiac surgery. (A–E) Patients undergoing cardiac surgery without cardiopulmonary bypass (no cardiopulmonary bypass [CPB]). (F–J) Patients undergoing CPB without the use of blood products in their priming fluids. (K–O) Patients undergoing CPB with fresh frozen plasma (FFP) in their priming fluids. (P–T) Patients undergoing CPB with red blood cells (RBCs) and FFP in their priming fluids. The concentration of individual phthalates (DEHP, MEHP, MECPP, MEHHP) and the sum phthalate equivalent level (DEHP + MEHP + MECPP + MEHHP) are reported. Matched values for each individual patient are connected by a line. Phthalate levels were measured preoperatively (preop), on postoperative day 0 (POD0), and at postoperative day 1 (POD1). Repeated measurements: normally distributed data analyzed via repeat measures ANOVA (equal variance) with Geisser–Greenhouse correction (unequal variance). Nonparametric data analyzed via Friedman test. FDR (0.1 cutoff) was used to correct for multiple comparisons; statistical significance denoted by *p < .05; **p < .01; ***p < .001; ****p < .0001; ns, not significant. [Color figure can be viewed at wileyonlinelibrary.com]

By POD1, the equivalent level declined to ±15% of the preoperative level in the “No CPB,” “Crystalloid prime,” and “Crystalloid prime + FFP” groups. However, there was a persistent elevation in MECPP, a secondary metabolite of DEHP, at this later time point relative to the preoperative measurements (No CPB: +45%, crystalloid prime + FPP: +306%; Figure 3c, m). In the RBC-based prime group, the equivalent level remained 147% higher at POD1 compared to preoperative baseline (Figure 3t). At POD1, DEHP levels returned to baseline, but primary and secondary metabolites remained elevated (MEHP: +309%, MECPP: +785%, MEHHP: +375%; Figure 3p–s). Since phthalate concentrations were comparable between CPB patients receiving either crystalloid-based prime or crystalloid-based prime + FFP, these patients were combined for further analyses.

3.3 |. Phthalate exposure varies by surgical intervention and age

Previous studies report increased DEHP exposure in NICU patients receiving multiple medical interventions that employ plastic devices.^22–25 Transfused patients can also be exposed to phthalates that leach into stored blood products.^37,58 We found that patients undergoing cardiac surgery without CPB had the lowest phthalate levels after surgery (1.38 ± 0.32 μM equivalent level; Figure 4a–e). Phthalate levels were higher in patients undergoing CPB with a crystalloid prime ± FFP (3.45 ± 1.97 μM equivalent level), which may be attributed to plastic materials in the CPB circuit. At POD0, MEHP was the most abundant metabolite detected in this group (1.42 ± 1.23 μM MEHP). Patients undergoing CPB with an RBC-based prime had the highest chemical exposure (11.40 ± 6.19 μM equivalent level). MEHP was also the most abundant metabolite detected in this group at POD0 (5.12 ± 4.37 μM MEHP). Since patient demographics varied between groups, we implemented multivariable linear regression to adjust for covariates that could influence phthalate exposure (Supplemental Table 2). As anticipated, patients undergoing CPB with an RBC-based prime had a significantly higher total phthalate exposure immediately after surgery (POD0), as compared to those undergoing cardiac surgery without CPB (p < 0.0001). We also sought to understand how quickly phthalates are cleared from patients’ blood after clinical exposure. All DEHP metabolites remained elevated at POD1 in patients undergoing CPB with an RBC-based prime (Figure 4, Supplemental Table 1).

FIGURE 4.

Higher phthalate chemical concentrations are detected in patients requiring cardiopulmonary bypass (CPB) with an red blood cell (RBC)-based prime. (A–E) DEHP, MEHP, MECPP, MEHHP, and the phthalate equivalent level (DEHP + MEHP + MECPP + MEHHP) in blood samples measured at postoperative day 0, and (F–J) postoperative day 1. For comparison, the mean preoperative value (all three groups combined) is indicated by a dashed line. Independent measurements analyzed via one-way ANOVA with Welch’s correction (unequal variance). FDR (0.1 cutoff) was used to correct for multiple comparisons; statistical significance denoted by *p < .05; **p < .01; ***p < .001; ****p < .0001; ns, not significant. [Color figure can be viewed at wileyonlinelibrary.com]

Next, we explored CPB variables that could influence chemical exposure as the equivalent level ranged broadly in CPB patients (0.77–31.57 μM, POD0). In young patients, underdeveloped glucuronidation and/or reduced renal function could diminish their ability to metabolize/excrete phthalates.⁵⁹ We found a significant correlation between age and phthalate concentration in patients receiving an RBC-based prime (r = −0.71, p < .0001), where infants <1 year old had increased phthalate levels (12.34 ± 6.0 μM) compared to patients >1 year (3.87 ± 2.05 μM equivalent level; Figure 5a). This trend persisted when age was compared to a weight- or BSA-corrected phthalate concentration (Supplemental Figure 1A,B). In vitro experiments suggest phthalate leaching from tubing is influenced by surface area, flow rate, temperature, perfusion time and/or storage duration.^{10,37,60–63} Since CPB flow rate is calculated by multiplying BSA with cardiac index,⁶⁴ flow rate in younger patients is inherently tied to BSA. Therefore, we measured the indexed flow rate (flow rate/BSA), which showed no correlation to phthalate exposure (Figure 5b). Absolute prime volume is also influenced by age, as younger patients receive a smaller CPB prime volume. We measured the prime volume indexed by BSA and found a positive correlation with phthalate levels (r = 0.67, p < .0001; Figure 5c).

FIGURE 5.

Surgical variables that may contribute to heightened phthalate exposure. (A) Phthalate levels were elevated in patients <1 year old. (B) Phthalate levels did not correlate with index flow rate. (C) Phthalate levels increased with index prime volume. (D–F) Phthalate levels increased with cardiopulmonary bypass (CPB) duration, and there was a weak correlation with venous temperature and fluid balance. (G–I) Phthalate levels positively correlated with increased indexed unwashed red blood cell (RBC) blood volume, indexed washed blood volume, and indexed fresh frozen plasma (FFP) volume in CPB priming fluids. (J) Patients <5 kg received washed RBCs (wRBC) and had higher phthalate blood levels. (K) Cell washing decreased DEHP and MEHP concentrations in an in vitro study using ≤8 day old blood units. (L) Heightened phthalate levels in patients receiving washed RBCs, may be influenced by administration of steroids (10 mg/kg methylprednisolone) that alter metabolism. Phthalate equivalent level (DEHP + MEHP + MECPP + MEHHP) reported on postoperative day 0 (POD0). (A–I) Spearman correlation indicated for each variable, with denoted p-value. (J, L) Two-tailed student’s t-test. (K) Two-tailed repeat measures t-test. *p < .05, ***p < .001, ****p < .0001. [Color figure can be viewed at wileyonlinelibrary.com]

DEHP leaching from tubing is also influenced by solution temperature and duration of contact,^60,63 with an approximate leaching rate of 3.12 g/L/h using a heart-lung machine.³⁹ In agreement, we found that longer CPB duration was associated with increased phthalate exposure (r = 0.54, p < .0001; Figure 5d). We anticipated that therapeutic hypothermia during CPB might reduce phthalate exposure but detected only a modest association (r = −0.25, p < .01; Figure 5e). This is likely due, in part, to patients who undergo longer CPB also undergoing therapeutic hypothermia. Finally, we found a modest correlation between patient fluid balance and postsurgical phthalate levels (r = 0.25, p < .01; Figure 5f).

Finally, we assessed whether phthalate exposure correlates with blood product usage as 91.9% of CPB patients received FFP, 29.7% received unwashed RBCs, and 51.4% received washed RBCs. Indexed volumes of unwashed RBCs (r = 0.48, p < .005), washed RBCs (r = 0.43, p < .001), and FFP (r = 0.69, p < .0001) in the CPB prime were positively correlated with POD0 phthalate levels (Figure 5g–i). These trends persisted when phthalate exposure was compared to the total volume of un/washed RBCs transfused during an operation (Supplemental Figure 1C,D). Unfortunately, we do not know which patients received blood products stored in DEHP versus DEHP-free bags (spot checking estimated >97% of RBC units and >70% of FFP units were manufactured with DEHP). Controlling for age (<1 year old), we compared phthalate levels in CPB patients who received either unwashed or washed RBCs and found that the latter had a 2.3-fold increase in phthalate levels (unwashed: 6.15 ± 3.7 μM, washed: 14.3 ± 5.4 μM equivalent level; Figure 5j). This result was unexpected, as cell washing removes extracellular contaminants from RBC units.⁶⁵ Indeed, our in vitro study showed that cell washing reduced DEHP levels in RBC units by 77% (pre-washing: 57.3 ± 25.5 μM, post-washing: 13.4 ± 4.9 μM DEHP) and MEHP levels by 74% (pre-washing: 2.9 ± 0.8 μM, post-washing: 0.75 ± 0.3 μM MEHP; Figure 5k). We speculate that higher phthalate levels in patients receiving washed RBCs might be attributed to perioperative administration of steroids, as 55.7% of patients receiving washed RBCs were also administered 10 mg/kg methylprednisolone. These patients had a 55% increase in phthalate levels compared to weight-matched controls (no steroids: 11.2 ± 4.4 μM, 10 mg/kg methylprednisolone: 17.4 ± 5.0 μM), suggesting that steroid effects on lipase activity may be influencing phthalate metabolism in this group.^66,67

3.4 |. Phthalate exposure and altered electrolyte balance

We investigated trends between phthalate exposure and postoperative blood gas parameters in CPB patients, as experimental work suggests that DEHP alters serum biochemistry.⁶⁸ All CPB cases were included in this analysis (n = 111) and normal laboratory values were determined by patient age. We measured higher phthalate levels in CPB patients who presented with hypokalemia (low K⁺: 12.5 ± 6.7 μM; normal K⁺: 8.3 ± 5.8 μM equivalent level), hypernatremia (high Na⁺: 11.7 ± 6.7 μM; normal Na⁺: 7.8 ± 5.5 μM equivalent level), and a higher anion gap (high anion gap: 12.9 ± 6.7 μM; normal anion gap: 7.0 ± 4.6 μM equivalent level; Figure 6a,b,h). Elevated fasting glucose levels have also been reported in individuals with higher daily phthalate exposure.⁶⁹ Similarly, we found elevated phthalate levels in patients with hyperglycemia (high glucose: 10.8 ± 6.9 μM; normal glucose: 8.1 ± 5.1 μM equivalent level; Figure 6d). Phthalate exposure was inversely associated with hypocalcemia (low calcium: 3.3 ± 2.2 μM; normal calcium: 9.8 ± 6.6 μM equivalent level), lower lactate levels (low lactate: 5.4 ± 3.9 μM; normal lactate: 10.5 ± 5.9 μM equivalent level), and lower pH (low pH: 3.3 ± 1.6 μM; normal pH: 11.1 ± 6.5 μM equivalent level; Figure 6c,g,i). We observed no significant difference in chloride or magnesium levels based on phthalate exposure (Figure 6e,f).

FIGURE 6.

Association between postoperative phthalate levels and clinical laboratory values in patients receiving cardiopulmonary bypass (CPB). Laboratory values are grouped into normal, low, and high ranges according to clinical standards for (A) potassium, (B) sodium, (C) calcium, (D) glucose, (E) chloride, (F) magnesium, (G) lactate, (H) anion gap, and (I) pH. Equivalent level (sum of DEHP + MEHP + MECPP + MEHHP) on postoperative day 0 (POD0) is reported on the y-axes. Statistical analysis by one-way ANOVA (three groups) with Welch’s correction for unequal variance, or two-tailed student’s ttest (two groups). *p < .05; **p < .01; ****p < .0001; ns, not significant; n/a, not applicable as patients were not observed in this group. [Color figure can be viewed at wileyonlinelibrary.com]

3.5 |. Heightened phthalate exposure is associated with postoperative complications

In experimental (rodent) models, phthalate exposure precipitates electrical disturbances leading to ventricular tachycardia, ventricular fibrillation, premature ventricular contractions, and cardiac arrest.^29,31,38 Accordingly, we investigated associations between phthalate exposure and 48-h postoperative outcomes in age-matched (<1 year) CPB patients. Phthalate levels were lower in patients without postoperative complications (6.1 ± 2.6 μM equivalent level), and 2–3× higher in patients experiencing cardiac arrhythmias or mechanical instability, including junctional ectopic tachycardia (16.4 ± 7.0 μM equivalent level), supraventricular tachycardias (13.5 ± 6.5 μM equivalent level), ventricular tachycardia/fibrillation (20.6 ± 2.6 μM equivalent level), and heart block (13.0 ± 6.4 μM equivalent level, Figure 7a–e). Increased phthalate levels were also associated with low cardiac output syndrome and/or cardiogenic shock (14.7 ± 6.3 μM equivalent level), cardiac arrest (12.0 ± 6.8 μM equivalent level), hypotension (13.3 ± 5.3 μM equivalent level), and blood pressure liability (18.0 ± 4.7 μM equivalent level) (Figure 7g–i,k). We did not find a link between phthalate exposure and hypertension (Figure 7j). Studies suggest phthalates are associated with kidney injury,^70,71 and we found a positive association between phthalate exposure (16.0 ± 9.5 μM equivalent level) and acute kidney injury and/or renal insufficiency (Figure 7l). Additionally, phthalates alter cellular metabolism^72–74 and we observed a positive association between phthalate exposure and lactic acidosis as well as hyperglycemia requiring intervention (15.5 ± 6 μM and 19.2 ± 8.2 μM equivalent level, respectively; Figure 7m,p). Finally, we noted that patients with elevated phthalate levels were more likely to require mechanical support (ECMO) and/or a postoperative transfusion (13.6 ± 6.5 μM and 12.9 ± 5.4 μM equivalent level, respectively; Figure 7q,r) within the first 48 h after surgery.

FIGURE 7.

Association between phthalate levels and 48-h postoperative outcomes in age-matched cardiopulmonary bypass patients (<1 year old). Associations between elevated plasma phthalate equivalent level (DEHP + MEHP + MECPP + MEHHP) on postoperative day 0 and postoperative complications including: (A) junctional ectopic tachycardia (JET), (B) supraventricular tachycardias (SVTs), (C) ventricular tachycardia or fibrillation (VT/VF), (D) premature ventricular contractions (PVCs), (E) heart block, (F) bradycardia, (G) low cardiac output syndrome (LCOS) and/or cardiogenic shock (CS), (H) cardiac arrest, (I) hypotension, (J) hypertension, (K) blood pressure (BP) fluctuations, (L) acute kidney injury (AKI) and/or renal insufficiency, (M) lactic acidosis, (N) respiratory acidosis, and (O) capillary leak syndrome. Association between elevated plasma phthalate levels and additional postoperative treatment, including (P) intervention for hyperglycemia, (Q) extracorporeal membrane oxygenation (ECMO), and (R) additional transfusion(s). Comps: complications. Statistical analysis by two-tailed student’s t-test (equal variance), with Welch’s correction (unequal variance). *p < .05; **p < .01; ***p < .001; ****p < .0001; ns, not significant. [Color figure can be viewed at wileyonlinelibrary.com]

4 |. DISCUSSION

We quantified iatrogenic phthalate plasticizer exposure in pediatric patients undergoing cardiac surgery. DEHP and its metabolites were detected in all patient samples, but higher phthalate levels were detected in patients undergoing CPB with an RBC-based prime and phthalate levels remained elevated the day after surgery. Higher phthalate levels were associated with younger patient age, longer CPB duration, greater indexed prime volume, fluid balance, and RBC-based CPB prime. Our in vitro study showed washing RBCs was an effective strategy toward reducing phthalate levels in the CPB prime. However, unexpectedly, we detected elevated phthalate levels in patients <5 kg who received washed RBCs—which may be attributed to the immature metabolic capacity of these younger patients and/or the co-administration of methylprednisolone which may alter lipase activity and phthalate metabolism.^67,75 We also identified associations between phthalate exposure and postoperative complications, including cardiac electrical instabilities, impaired contractile function, blood pressure disturbances, and abnormal blood gas measurements.

In agreement with prior studies, our work identifies cardiac surgery and circulatory support as clinical sources of phthalate exposure. Barry et al. reported elevated blood DEHP levels in infants undergoing CPB, ranging from 1.1 to 5.6 μg/mL (n = 7).⁵⁰ Gaynor et al. also found increased urinary DEHP metabolite levels in infants after cardiac surgery (n = 18).⁷⁶ Eckert et al. reported increased blood phthalate levels postoperatively in infants undergoing CPB (n = 21), despite the use of DEHP-free tubing. Specifically, phthalate exposure was attributed to DEHP accumulation in stored RBC and FFP units.⁴⁸ Finally, Karle et al. reported elevated plasma phthalate concentrations (0–24.2 μg/mL DEHP) in infants receiving ECMO (n = 18).⁶² Importantly, previous work has demonstrated phthalates can localize to sensitive organs – including the heart tissue of infants who received umbilical catheters and/or blood transfusions prior to death.^26,27 Our work expands on these reports, benefitting from a larger study enrollment across a wider age range, which can help to better understand phthalate exposure and metabolism in pediatric patients. Moreover, our study includes a control group of cardiac surgery patients undergoing a thoracotomy without CPB, which helps to distinguish exposures from cardiac surgery alone versus components of the CPB circuit. Finally, we show that cell washing removes DEHP from fresh RBC units, building upon the work by Mϋnch et al. that demonstrated washing RBC units decreases DEHP and MEHP levels by 18%–96% in 36–56 day old blood units.⁴¹

Clinical studies linking phthalate exposure to patient outcomes are sparse, but experimental work suggests that phthalates exert cardiotoxic effects.^2,38,77,78 DEHP exposure halted spontaneous beating activity of chick cardiomyocytes¹² and slowed heart rate, atrioventricular conduction, and impaired contractile activity in rodent models.^29,30,79 Further, MEHP infusion can precipitate hypotension and cardiac arrest in rodents.³¹ These experimental findings support the postoperative complications observed in this study, wherein CPB patients <1 year with heightened phthalate exposure were more likely to experience cardiac arrest, postoperative arrhythmias, and reduced contractile function. Taken together, it is plausible that a combination of risk factors (young age, longer CPB duration, increased phthalate exposure) collectively contribute to these complications.

Previously, the United States Food and Drug Administration issued a recommendation to minimize phthalate exposure in neonatal boys due to reproductive toxicity concerns.⁸⁰ This prompted NICUs to adopt alternative plastic products^81,82; yet the same efforts have not been applied to pediatric cardiac intensive care units or cardiothoracic surgery. Additional work is needed to investigate mitigation strategies to reduce DEHP exposure; this may include adopting alternative tubing materials less prone to leaching, using DEHP-free RBC storage bags, cell washing to remove phthalate contaminants, or decreasing CPB prime volumes. Evaluating these strategies is critical—as roughly 67% of operations reported in the Society of Thoracic Surgeons Congenital Heart Surgery Database occur within the first year of life.⁸³ Furthermore, cardiac surgery and perioperative intensive care can result in exceedingly high phthalate exposures for vulnerable neonates and children and these exposures may contribute to increased morbidity/mortality.

Limitations of our study include a heterogeneous study population, as the enrolled patients differ in age, complexity of their congenital heart defects, and surgical repair. Therefore, we cannot prove causation between phthalate exposure and postoperative complications, and additional work will be required to evaluate these associations further. One strength of our study is the inclusion of a “control” surgery group, which had limited exposure to DEHP-containing plastic products. Although it should be noted that this group had a limited sample size (n = 11), of which all patients were <1 year old, and each patient required a less complex surgery that did not necessitate CPB. As such, this could limit the extrapolation of our control data to older patients. It is also important to note that some patients could have received blood products (RBCs or FFP) that were stored in DEHP-free storage bags. DEHP-free bags are used much less commonly (spot checking estimated >97% of RBC units and >70% of FFP units used at Children’s National were manufactured with DEHP), but this could lead to considerable variations in postoperative phthalate blood levels. Our study also used blood samples that were drawn in accordance with normal clinical care, therefore, we could not standardize the exact time of sample collection. Peak blood phthalate concentrations are likely to occur during or immediately after surgery, yet the earliest time point we measured (POD0) coincided with arrival to the cardiac intensive care unit. POD1 samples were collected the morning after surgery, which could lead to slight variations in the length of time between the end of surgery and sample collection. Finally, postoperative outcomes were assessed retrospectively and were limited to the thoroughness of daily progress notes.

Abbreviations:

AKI

acute kidney injury

BP

blood pressure

BSA

body surface area

CICU

cardiac intensive care unit

CPB

cardiopulmonary bypass

CS

cardiogenic shock

DEHP

di(22010ethylhexyl) phthalate

ECMO

extracorporeal membrane oxygenation

ET

endotracheal tube

FDR

false discovery rate

FFP

fresh frozen plasma

JET

junctional ectopic tachycardia

LCOS

low cardiac output syndrome

LOS

length of stay

MECPP

mono(2-ethyl-5-carboxypentyl) phthalate

MEHHP

mono(2-ethyl-5-hydroxyhexyl) phthalate

MEHP

mono(2-ethylhexyl) phthalate

NICU

neonatal intensive care unit

PICU

pediatric intensive care unit

POD0

postoperative day 0

POD1

postoperative day 1

Preop

preoperative

PVC

polyvinyl chloride

PVCs

premature ventricular contractions

RBC

red blood cell

SVTs

supraventricular tachycardias

VF

ventricular fibrillation

VT

ventricular tachycardia

wRBCs

washed red blood cells