In a 2‑year‑old with severe ARFID who is receiving polyethylene glycol (PEG) for constipation and has developed new motor tics and worsening aggression, is there a medically reasonable basis for a supervised trial off PEG, what evidence supports or refutes an association, what alternative maintenance laxatives (including lactulose, magnesium hydroxide, and magnesium citrate) are appropriate, where does magnesium citrate fall outside guideline‑supported chronic use, what pediatric safety, dosing, and monitoring considerations apply, and what information do the pediatric gastroenterologist and developmental pediatrician need before agreeing to such a trial?

Is There a Medically Reasonable Basis for a Supervised Trial Off PEG?

Yes, there is a medically reasonable basis to propose a supervised dechallenge trial, but the evidence linking PEG to motor tics or aggression is essentially nonexistent, making this primarily a clinical judgment call to rule out an idiosyncratic reaction rather than a known drug effect. The decision framework should weigh the very low probability of PEG causation against the diagnostic value of a dechallenge in a child with severe ARFID where medication tolerability is already compromised.

Evidence For and Against PEG Causing Neuropsychiatric Symptoms

Against an Association (Stronger Evidence)

• PEG has an excellent safety profile in pediatric populations with only minor adverse events reported across multiple systematic reviews 1, 2, 3. The 2025 Lancet meta-analysis of 59 RCTs representing 7,045 children found PEG to be safe with moderate-to-high certainty evidence 3.
• Documented adverse events are gastrointestinal: bloating, flatulence, diarrhea, and abdominal pain 4. Neuropsychiatric effects are not mentioned in any guideline or systematic review.
• The 2023 AGA-ACG guideline found no conclusive evidence of serious adverse events with PEG (RR 0.47, CI 0.16–1.33), though the confidence interval is wide 4.
• PEG is minimally absorbed systemically—it acts as an osmotic agent in the gut lumen 4.

For Considering a Trial (Weaker but Clinically Relevant)

• Rare anaphylaxis to PEG has been documented 5, proving PEG is not completely inert and can trigger immune-mediated reactions in susceptible individuals.
• Absence of evidence is not evidence of absence: No large-scale studies have specifically examined neuropsychiatric outcomes with PEG in children with neurodevelopmental conditions.
• Temporal association matters clinically: If symptoms emerged or worsened after PEG initiation, a dechallenge provides diagnostic information even if the prior probability is low.

Alternative Causes to Consider Before Attributing to PEG

For Motor Tics

• Primary tic disorder emergence (typical onset 4-6 years, but can occur earlier)
• Developmental progression of underlying neurodevelopmental condition
• Other medications or supplements (even seemingly benign ones)
• Sleep deprivation or stress related to constipation/feeding difficulties
• PANDAS/PANS (if acute onset with temporal infection)
• Nutritional deficiencies (iron, magnesium, vitamin D) related to ARFID

For Worsening Aggression

• Uncontrolled pain from constipation itself (abdominal pain, anal fissures)
• Communication frustration in a child with severe ARFID who cannot express discomfort
• Behavioral escalation from repeated medication administration (taste/texture aversion)
• Sleep disruption from gastrointestinal discomfort
• Developmental regression or progression of underlying condition
• Environmental stressors or changes in routine

Clinical Value of a Dechallenge

A supervised trial off PEG would provide useful clinical information if:

• Symptoms resolve or significantly improve within 2-4 weeks off PEG (suggesting possible association)
• Symptoms persist unchanged (ruling out PEG, redirecting diagnostic workup)
• Constipation worsens significantly (confirming PEG efficacy, informing risk-benefit of alternatives)

The dechallenge is most informative when:

• Baseline symptom severity is documented with validated scales (e.g., YGTSS for tics, standardized aggression measure)
• Alternative laxative is equally effective (avoiding confounding from worsening constipation)
• Duration is sufficient (minimum 4 weeks to see neuropsychiatric changes)
• Reintroduction is planned if symptoms resolve (confirming causation)

---

Realistic Maintenance Alternatives for a 2-Year-Old with Severe ARFID

First-Line Guideline-Supported Options

Magnesium Oxide (MgO)

This is the most practical alternative given the taste/texture sensitivities.

• Guideline support: The 2023 AGA-ACG guideline suggests MgO for chronic constipation (conditional recommendation, very low certainty) 4.
• Pediatric evidence: A 2023 RCT found MgO equally effective as PEG in children 6 months to 18 years, with better acceptance in children under 4 years 6. A 2011 study confirmed equal efficacy but noted 42.9% refused magnesium hydroxide due to taste 7.
• Practical advantage: MgO is typically available as small tablets or capsules that can be opened and mixed into food, potentially easier to disguise than liquid formulations.
• Dosing: Adult trials used 1.5 g/day 4; clinical practice often uses 500 mg/day to 1 g/day 4. For a 2-year-old, start at 250-500 mg/day divided into 1-2 doses, titrate based on response.
• Monitoring: Serum magnesium and creatinine at baseline and every 3-6 months; avoid if creatinine clearance <20 mL/dL 4.
• Key limitation: Only MgO has RCT data; other magnesium formulations (including citrate) lack evidence 4.

Lactulose

Less ideal due to taste, but guideline-supported.

• Guideline support: 2023 AGA-ACG suggests lactulose for adults who fail OTC therapies (conditional recommendation, very low certainty) 4.
• Pediatric dosing: FDA label states 2.5-10 mL/day for infants in divided doses 8. For a 2-year-old, start 5 mL once or twice daily, adjust to produce 2-3 soft stools daily.
• Practical disadvantage: Sweet taste and syrup consistency are often rejected by children with sensory sensitivities. Bloating and flatulence are dose-dependent and common 4.
• Evidence: A 2010 Cochrane review found PEG superior to lactulose for stool frequency, consistency, and abdominal pain relief 9. A 2014 meta-analysis confirmed PEG better tolerated 1.

Magnesium Hydroxide (Milk of Magnesia)

Likely to be rejected based on your description.

• Pediatric evidence: Equal efficacy to PEG but significantly worse acceptance, especially in children >4 years 6, 7.
• Practical issue: Chalky texture and taste make it unsuitable for a child with severe ARFID and medication sensitivities.

---

Where Magnesium Citrate Falls Outside Guidelines

Guideline Position

Magnesium citrate is explicitly NOT supported by current guidelines for chronic maintenance therapy.

The 2023 AGA-ACG guideline states: "Only MgO has been evaluated in RCTs; the bioavailability and clinical efficacy of other formulations of magnesium (eg, citrate, glycinate, lactate, malate, sulfate) for CIC are unknown" 4.

Why This Matters

• Zero RCT evidence for magnesium citrate in chronic constipation (pediatric or adult)
• Different pharmacokinetics: Citrate formulation has different absorption and osmotic properties than oxide
• Off-label use: Any use would be entirely empiric, not evidence-based
• Regulatory status: Magnesium citrate is FDA-approved as a saline laxative for bowel preparation, not chronic maintenance

Pediatric Safety and Dosing Considerations for Magnesium Citrate

What Is Actually Known (Very Little)

• No pediatric RCTs for chronic constipation maintenance
• No established chronic dosing regimens in children
• Acute use data only: Used for bowel preparation at single doses of 2-4 mL/kg (maximum 300 mL)
• Absorption differences: Citrate is more bioavailable than oxide, increasing hypermagnesemia risk

Theoretical Risks in a 2-Year-Old

1. Hypermagnesemia: Higher absorption than MgO increases risk, especially with:

   • Renal immaturity (though less concerning at age 2 than in neonates)
   • Chronic daily dosing (no safety data)
   • Dehydration (common in ARFID with fluid aversion)

2. Electrolyte disturbances: Chronic osmotic load may affect sodium, potassium, and fluid balance

3. Diarrhea and dehydration: More potent osmotic effect than MgO may cause excessive stooling

4. Drug interactions: FDA label warns to take 2+ hours apart from other drugs 10

5. Lack of monitoring guidelines: No established parameters for chronic pediatric use

Required Monitoring If Considered

• Baseline: Serum magnesium, comprehensive metabolic panel (electrolytes, BUN, creatinine), urinalysis
• Ongoing: Repeat labs every 2-4 weeks initially, then monthly if stable
• Clinical: Daily stool frequency/consistency, hydration status, signs of hypermagnesemia (hypotonia, lethargy, decreased reflexes, respiratory depression)
• Dose: Would need to extrapolate from MgO dosing (start very low, e.g., 50-100 mg elemental magnesium daily)

---

Questions the Pediatric GI and Developmental Pediatrician Need Answered

For the Pediatric Gastroenterologist

Regarding the Dechallenge

1. What is the current constipation severity and PEG dose? (Frequency, consistency, impaction, pain)
2. What was the temporal relationship between PEG initiation and symptom onset? (Days? Weeks? Months?)
3. Has PEG dose been adjusted recently? (Dose changes can affect behavior via GI symptoms)
4. What is the plan for constipation management during the trial? (Which alternative, at what dose, with what monitoring?)
5. What is the threshold for restarting PEG? (How many days without stool? Impaction? Pain?)
6. What is the family's ability to monitor and document? (Stool diary, symptom log)

Regarding Alternative Laxatives

1. Has the child tried other laxatives previously? (Response, tolerance, reasons for discontinuation)
2. What formulations have been attempted for medication administration? (Mixed in food, specific vehicles, timing)
3. Is there evidence of malabsorption or other GI pathology? (Would affect magnesium absorption/risk)
4. What is the renal function? (Creatinine, eGFR—critical for any magnesium-based therapy)

Regarding Magnesium Citrate Specifically

1. Why is magnesium citrate being considered over evidence-based MgO? (Is there a specific rationale, or is it availability/formulation-driven?)
2. What is the proposed dosing regimen? (Starting dose, titration plan, maximum dose)
3. What is the monitoring plan? (Lab frequency, clinical parameters, duration of trial)
4. What are the stopping criteria? (Hypermagnesemia threshold, inefficacy definition, adverse events)

For the Developmental Pediatrician

Regarding the Neuropsychiatric Symptoms

1. When exactly did the tics first appear? (Date, in relation to PEG start, any other changes)
2. What is the tic phenomenology? (Motor vs. vocal, simple vs. complex, frequency, suppressibility)
3. Has there been formal tic assessment? (YGTSS score or equivalent)
4. What is the aggression pattern? (Frequency, triggers, severity, change from baseline)
5. Has there been formal behavioral assessment? (Standardized scales, functional analysis)
6. Are there other new symptoms? (Sleep changes, OCD symptoms, anxiety, regression)

Regarding Differential Diagnosis

1. What is the underlying neurodevelopmental diagnosis? (Autism, ADHD, intellectual disability, genetic syndrome)
2. Is there a family history of tic disorders or OCD? (Genetic predisposition)
3. Have there been recent infections? (PANDAS/PANS consideration)
4. What other medications or supplements is the child taking? (Including vitamins, probiotics, herbal remedies)
5. What is the nutritional status? (Specific deficiencies that could cause or worsen symptoms)
6. What environmental or psychosocial stressors are present? (Changes in routine, school, family)

Regarding the Trial Design

1. What is the baseline symptom severity using validated measures? (Not just parental report)
2. What is the planned duration of the dechallenge? (Sufficient to see neuropsychiatric change)
3. Who will monitor neuropsychiatric symptoms during the trial? (Frequency, tools, thresholds for concern)
4. Is there a plan for reintroduction if symptoms resolve? (To confirm causation vs. coincidence)

---

Flaws and Unsafe Assumptions in Your Reasoning

Potential Flaws

1. Overweighting Temporal Association

• Flaw: Assuming temporal correlation implies causation when the prior probability of PEG causing these symptoms is extremely low.
• Correction: Frame this as "ruling out a rare idiosyncratic reaction" rather than "investigating a likely drug effect."

2. Underestimating Constipation as a Cause of Aggression

• Flaw: Not fully considering that untreated or undertreated constipation itself causes pain, discomfort, and behavioral changes in nonverbal or minimally verbal children.
• Correction: Ensure the alternative laxative is equally effective, or the dechallenge will be confounded by worsening constipation.

3. Considering Magnesium Citrate Without Evidence

• Flaw: Proposing an intervention with zero evidence base when evidence-based alternatives (MgO) exist.
• Correction: If taste/texture is the issue, focus on creative administration strategies for MgO (compounding, mixing vehicles) rather than switching to an unstudied formulation.

4. Inadequate Monitoring Plan for Magnesium-Based Therapy

• Flaw: Not specifying the frequency and parameters for monitoring hypermagnesemia in a 2-year-old with potential fluid intake issues (ARFID).
• Correction: Establish explicit lab monitoring schedule (baseline, 2 weeks, 4 weeks, then monthly) and clinical signs to watch for.

5. Insufficient Baseline Documentation

• Flaw: Planning a dechallenge without validated baseline measures of tics and aggression makes it impossible to objectively assess change.
• Correction: Require formal assessment (e.g., YGTSS for tics, ABC or CBCL for aggression) before and during the trial.

Unsafe Assumptions

1. Assuming ARFID Won't Worsen with Medication Changes

• Risk: Introducing a new laxative with different taste/texture could trigger further food/medication refusal, worsening both ARFID and constipation.
• Mitigation: Involve feeding therapist in medication administration planning; consider compounded formulations.

2. Assuming Renal Function Is Normal

• Risk: Any magnesium-based therapy in a child with potential dehydration (ARFID) requires confirmed normal renal function.
• Mitigation: Obtain baseline creatinine and eGFR before starting any magnesium therapy.

3. Assuming Symptoms Will Clearly Resolve or Persist

• Risk: Neuropsychiatric symptoms in young children with neurodevelopmental conditions fluctuate naturally; a 4-week trial may not provide clear answers.
• Mitigation: Plan for longer trial (8-12 weeks) with multiple assessment points and consider reintroduction challenge.

4. Assuming Family Can Manage Complex Monitoring

• Risk: Expecting detailed symptom logs, stool diaries, and recognition of hypermagnesemia signs may exceed family capacity, especially with a child with severe ARFID.
• Mitigation: Provide structured logs, frequent check-ins (weekly initially), and clear red flags for seeking immediate care.

5. Assuming GI and Developmental Teams Will Coordinate

• Risk: Without explicit communication plan, monitoring may fall through cracks (e.g., GI assumes developmental is tracking behavior; developmental assumes GI is tracking labs).
• Mitigation: Designate one physician as primary coordinator, establish shared documentation system, schedule joint case conference at trial midpoint.

---

Practical Algorithm for Decision-Making

Step 1: Establish Baseline (Before Any Changes)

• Formal tic assessment (YGTSS or equivalent)
• Formal aggression/behavior assessment (ABC, CBCL, or equivalent)
• Constipation severity (Rome IV criteria, stool diary for 1 week)
• Labs: CBC, CMP (including magnesium), creatinine, eGFR
• Document all current medications, supplements, and recent changes

Step 2: Determine If Dechallenge Is Justified

Proceed if:

• Clear temporal relationship (symptoms began or worsened within 4 weeks of PEG start/dose increase)
• No other obvious cause identified (infection, medication change, major stressor)
• Family understands low probability of PEG causation but wants diagnostic clarity
• Alternative laxative plan is in place

Do NOT proceed if:

• Symptoms preceded PEG or have been stable for months on PEG
• Constipation is poorly controlled (dechallenge will be confounded)
• Family cannot commit to monitoring or follow-up
• Renal function is abnormal

Step 3: Select Alternative Laxative

First choice: Magnesium oxide

• Start 250 mg daily (or 125 mg twice daily) for a 2-year-old
• Titrate by 125-250 mg every 3-5 days based on stool output
• Target: 2-3 soft stools daily without pain or straining

Second choice: Lactulose

• Start 5 mL once daily
• Titrate by 2.5-5 mL every 3-5 days
• Accept that taste may be an issue; consider mixing strategies

Avoid: Magnesium citrate (no evidence, higher risk)

Step 4: Implement Dechallenge with Monitoring

Week 0 (Transition week):

• Stop PEG
• Start alternative laxative at low dose
• Daily stool diary
• Daily brief behavior log (tic frequency, aggression episodes)

Weeks 1-4:

• Continue stool and behavior logs
• Labs at week 2 (magnesium, electrolytes, creatinine)
• Phone check-in weekly
• Adjust laxative dose as needed for constipation control

Weeks 5-8:

• Formal tic reassessment (YGTSS)
• Formal behavior reassessment (ABC/CBCL)
• Labs at week 8
• In-person visit to review data

Step 5: Interpret Results and Decide Next Steps

If symptoms improved AND constipation controlled:

• Continue alternative laxative
• Consider PEG reintroduction challenge in 4-8 weeks to confirm causation (if family willing)

If symptoms unchanged AND constipation controlled:

• PEG ruled out as cause
• Continue alternative laxative if well-tolerated
• Pursue other diagnostic workup for tics/aggression

If symptoms improved BUT constipation worsened:

• Unclear result (improvement may be from reduced GI distress)
• Optimize alternative laxative dose
• Reassess in 4 weeks

If symptoms worsened OR constipation worsened:

• Resume PEG
• Pursue other causes for neuropsychiatric symptoms
• Consider GI and developmental co-management for symptom optimization

---

Bottom Line for the Case Conference

The proposal to trial off PEG is medically reasonable as a diagnostic maneuver, not because there is evidence linking PEG to tics or aggression (there is none), but because the clinical context—a child with severe ARFID and new concerning symptoms—justifies ruling out an idiosyncratic reaction. The key is ensuring the alternative laxative (preferably MgO, NOT magnesium citrate) is equally effective, monitoring is rigorous, and both teams understand this is a low-probability investigation that may not yield clear answers. Magnesium citrate should be avoided unless MgO and lactulose both fail, and only then with explicit informed consent about the lack of evidence and enhanced monitoring protocol.