Topical NSAID exposure
Evidence for a focused reference app

Clinical review draft · Canada first · Sources checked October 23, 2026 · APA 7 references

A useful first version can compare exact studied formulations and regimens. A body-area picker can help select treated sites; the evidence does not establish a knee-to-ribs exposure multiplier.

This is the requested evidence and design review, not an implemented app or dosing calculator. Reported measures and units are preserved as published. Calculations and unknowns are identified separately.

Scope and decisions to review

Start with Canadian diethylamine gel in its two strengths and Canadian Pennsaid solution. Give InflaFlex an information page with its evidence gaps visible. Keep international gels, ketoprofen studies and historical formulations in an expandable evidence library. Brand variants sharing a monograph can share evidence without treating different salts or vehicles as interchangeable.

Canadian dosing and formulation inventory

Label dosing is separate from research dosing. Every dose below links to its source; the detailed study records also retain application conditions and unknowns.

Exact formulationCanadian label directionsPopulationVehicle
Voltaren Emulgel 1.16%

diclofenac diethylamine · 1.16% w/w

2–4 g for approximately 400–800 cm², 3–4 times daily; no more than 4 applications in 24 h. Beyond 7 days only if recommended by a doctor. (Haleon Canada ULC, 2026a, §§4.2, 4.4, p. 5)Adults and adolescents ≥16 years; usual adult dosage in geriatrics. (Haleon Canada ULC, 2026a, §1)Carbomer, cocoyl caprylocaprate, diethylamine, fragrance (containing benzyl benzoate), isopropyl alcohol, liquid paraffin, macrogol cetostearyl ether, propylene glycol, purified water. (Haleon Canada ULC, 2026a, §6, pp. 7–8)
Voltaren Emulgel Extra Strength 2.32%

diclofenac diethylamine · 2.32% w/w

2 g in the morning and 2 g in the evening; maximum 4 g/day over ALL affected areas. Beyond 7 days only if recommended by a doctor. (Haleon Canada ULC, 2026b, §4.2, p. 5)Indication for adults 18–65 years; no authorized indication for >65 years or <18 years. This is distinct from a contraindication. (Haleon Canada ULC, 2026b, §1)Butylhydroxytoluene, carbomer, cocoyl caprylocaprate, diethylamine, fragrance, isopropyl alcohol, liquid paraffin, macrogol cetostearyl ether, oleyl alcohol, propylene glycol, purified water. (Haleon Canada ULC, 2026b, §6, p. 8)
Pennsaid 1.5%

diclofenac sodium · 1.5% w/w

50 drops per knee 3 times daily OR 40 drops per knee 4 times daily; maximum treatment duration 3 months, continuous or intermittent. (Knight Therapeutics Inc., 2025, §4.2, Table 1, p. 6)Adults ≥18; pediatric use contraindicated. Lower starting dose may be considered for older/frail patients under close supervision; no numerical reduction supplied. (Knight Therapeutics Inc., 2025, §1)Dimethyl sulfoxide (DMSO), ethanol, glycerine, propylene glycol, purified water. The current Canadian composition table does not quantify DMSO. (Knight Therapeutics Inc., 2025, §6, p. 8)

Studied exposure: reported values

Each row stays within its own study. Values from different rows should not be ranked as though the populations, sites, doses and sampling conditions were matched. “±” and ranges retain the source’s summary method. Source percentages are not recalculated or standardized.

Formulation / studyStudied applicationActual comparatorReported systemic exposurePercentage meaning / limit
Canadian diethylamine gel 1.16%
NCH study; healthy volunteers		4 g three times daily; 7 days. Site and area not reported. (Haleon Canada ULC, 2026a, p. 26, Table 4)Voltaren enteric-coated 25 mg three times daily. (Haleon Canada ULC, 2026a, p. 26)Day 7 geometric means (range):
AUC₀–₂₄ 179 (51.8–332) ng·h/mL
Cmax 12.0 (2.54–45.1) ng/mL.
Oral AUC 1360 (327–4690) ng·h/mL. (Haleon Canada ULC, 2026a, Table 4)		Reported rounded AUC comparison: 13% of oral at last administration; not dose-normalized. (Haleon Canada ULC, 2026a, p. 26)
Canadian diethylamine gel 2.32%
VOPO-PE-102; healthy volunteers		2 g twice daily to approximately 400 cm² on the same ankle; 7 days; non-occlusive. (Haleon Canada ULC, 2026b, pp. 26–28)Diclofenac sodium 50 mg three times daily; tablet release type not specified in summary. (Haleon Canada ULC, 2026b, p. 26)Day 7 geometric means (range):
AUC₀–₂₄ 67.9 (38.2–255.2) ng·h/mL
Cmax 4.6 (2.2–29.4) ng/mL.
Oral AUC 2830 (249–5955) ng·h/mL. (Haleon Canada ULC, 2026b, Table 6)		Reported 4.5% relative bioavailability is dose-normalized to an equivalent diclofenac sodium oral dose. It is not the percentage of applied gel absorbed. (Haleon Canada ULC, 2026b, §10.3)
Canadian diethylamine gel 1.16%
VOPO topical comparator		2 g four times daily to approximately 400 cm² on the same ankle; 7 days; non-occlusive. (Haleon Canada ULC, 2026b, pp. 26–27)Direct comparator here is topical 2.32%, not a separate oral arm displayed in Table 4. (Haleon Canada ULC, 2026b, Table 4)Day 7 geometric means:
AUC₀–₂₄ 62.4 ng·h/mL
Cmax 4.6 ng/mL. (Haleon Canada ULC, 2026b, Table 4)		Reported extra/regular AUC ratio 108.4% (90% CI 92.9–126.5). This is a topical-to-topical comparison. (Haleon Canada ULC, 2026b, Table 4)
Pennsaid 1.5% solution
Holt Study 2; 22 completers		40 drops (approximately 1.2 mL) to each knee four times daily; 7.5 days. Area in cm² not reported. (Holt et al., 2015, Methods)Diclofenac sodium 75 mg twice daily; methods say delayed-release, Table 4 says sustained-release. (Holt et al., 2015, Methods; Table 4)Steady-state day 8 mean (SD):
AUCss₀–₂₄ 353.78 (174.93) ng·h/mL
Cmaxss 18.07 (8.99) ng/mL.
Oral AUC 4426.01 (2793.74) ng·h/mL. (Holt et al., 2015, Table 4)		Retain AUC as reported. Do not borrow the published 2%/oral percentage for the 1.5% arm. Oral release identity needs resolution.
International and historical formulations
FormulationStudied regimenReported resultInterpretation
Diclofenac sodium gel 1%, US4 g to one knee four times daily, 400 cm², 7 days; oral comparator 50 mg three times daily. (Kienzler et al., 2010, Abstract)AUC₀–₂₄ 233 ± 128 ng·h/mL; Cmax 15 ± 7.3 ng/mL. Reported AUC 5.8% of oral (CI 5–6.7; CI level not stated in label table). (Rebel Distributors Corp, 2010, §12.3, Table 2)Distinct from Canadian Emulgel. Historical label used to supplement original-study abstract.
Diclofenac sodium gel 1%, US, larger-area armTwo knees and two hands: 48 g/day, 1200 cm², 7 days; same oral comparator. (Kienzler et al., 2010, Abstract)AUC₀–₂₄ 807 ± 478 ng·h/mL; Cmax 53.8 ± 32 ng/mL. Reported AUC 19.7% of oral (CI 17–22.8). (Rebel Distributors Corp, 2010, §12.3, Table 2)Research regimen exceeds the historical 32 g/day label maximum; not a suggested dose. (Rebel Distributors Corp, 2010, §2.3)
Pennsaid 2% solution, Holt Study 22 mL to each knee twice daily, 7.5 days; same oral arm as 1.5% study. (Holt et al., 2015, Methods)Day 8 mean (SD) AUCss₀–₂₄ 315.47 (158.70) ng·h/mL; Cmaxss 19.53 (10.31) ng/mL. Reported least-squares-mean AUC ratio 7.75% (90% CI 6.07–9.90). (Holt et al., 2015, Table 4)Do not replace the reported adjusted ratio with a ratio of arithmetic means. Current 2% labeling is outside the Canadian MVP.
Ketoprofen 2.5% gel; brand/vehicle unconfirmed15 g/day (375 mg ketoprofen) on back, 750 cm², for 10 days. (Flouvat et al., 1989, Abstract)First-dose Cmax 144 ± 91 ng/ml. Urinary recovery about 2.6% of first applied dose. No accumulation reported; no quantitative AUC supplied. (Flouvat et al., 1989, Abstract)No oral comparator; no exposure ratio. Do not identify this as Tiloket without formulation confirmation.
Ketoprofen 20% PLOSingle topical 200 mg ketoprofen versus oral 50 mg capsule; blood sampling through 10 h. (Dowling et al., 2004, Abstract)Median Cmax 0.021 µg/mL topical vs 4.15 µg/mL oral. Reported median relative bioavailability 0.48% (range 0.18–2.1%). (Dowling et al., 2004, Abstract)Original units retained. Full text needed to verify the relative-bioavailability formula and AUC interval; not transferable to InflaFlex.
Historical Pennsaid 1.5% radiolabeled lotionKnee application left for 24 h, then washed; amount and sample size unavailable in abstract. (Hui et al., 1998, Abstract)Reported topical absorption 6.6% of applied dose; peak plasma 14C at 30 h. (Hui et al., 1998, Abstract)Radiolabel/metabolite measurement; not interchangeable with parent-drug plasma AUC or with the monograph’s urinary recovery.
Full extraction details: variability, Tmax, eligibility and missing fields

The reusable JSON contains these records, source locators, original units and separately marked arithmetic. Missing fields are explicitly null with a reason.

NCH 17727B-510.20-02-01B

Application
4 g three times daily (Haleon Canada ULC, 2026a, §14.2, p. 26, Table 4)
Population
Healthy volunteers; randomized open 3-way crossover; {'enrolled': 24, 'topical': 24, 'oral': 23}
Renal eligibility
Healthy volunteers; detailed renal exclusion threshold unavailable
Conditions
skin: not reported in summary; heat: not reported; occlusion: not reported
Timing
7 days; final morning administration on day 7
AUC
179 ng·h/mL; geometric mean; range 51.8–332 (Haleon Canada ULC, 2026a, p. 26, Table 4)
Cmax
12 ng/mL; geometric mean; range 2.54–45.1 (Haleon Canada ULC, 2026a, p. 26, Table 4)
Tmax
18 h; median; range 0–20 (Haleon Canada ULC, 2026a, p. 26, Table 4)
Limitations
Do not call this a knee regimen: site and area are not reported. The monograph reports rounded AUC comparison of 13% at the last administration; unequal applied and oral daily doses. Day-7 wording is last morning administration; do not relabel as a validated 24-hour routine dosing prediction.

VOPO-PE-102

Application
2 g twice daily (Haleon Canada ULC, 2026b, §14.2, pp. 26–28, Tables 4 and 6)
Population
Healthy men and women; randomized open crossover; {'enrolled': 38, 'topical': 38, 'oral': 38}
Renal eligibility
Healthy volunteers; detailed renal exclusion threshold unavailable
Conditions
skin: intact skin context; no diseased-skin cohort; heat: not reported; occlusion: non-occlusive arm
Timing
7 days; day-7 assessment
AUC
67.9 ng·h/mL; geometric mean; range 38.2–255.2 (Haleon Canada ULC, 2026b, p. 28, Table 6)
Cmax
4.6 ng/mL; geometric mean; range 2.2–29.4 (Haleon Canada ULC, 2026b, p. 28, Table 6)
Tmax
19 h; median; range 0–24 (Haleon Canada ULC, 2026b, p. 28, Table 6)
Limitations
4.5% is dose-normalized relative bioavailability, not the raw AUC ratio and not percentage of applied dose absorbed. Semi-occlusive arm was also studied, but the main values here are non-occlusive. Do not transfer ankle data to ribs or scale by body surface area.

VOPO-PE-102, regular-strength comparator

Application
2 g four times daily (Haleon Canada ULC, 2026b, §14.2, pp. 26–27, Table 4)
Population
Healthy men and women; same crossover study; {'enrolled': 38}
Renal eligibility
Healthy volunteers; detailed threshold unavailable
Conditions
skin: intact skin context; heat: not reported; occlusion: non-occlusive
Timing
7 days
AUC
62.4 ng·h/mL; geometric mean (Haleon Canada ULC, 2026b, p. 27, Table 4)
Cmax
4.6 ng/mL; geometric mean (Haleon Canada ULC, 2026b, p. 27, Table 4)
Tmax
Not reported in retrieved source (Haleon Canada ULC, 2026b, p. 27, Table 4)
Limitations
Direct topical-to-topical comparison is supported. Oral AUC for this arm is not displayed in Table 4. Table 4 supplies arithmetic Tmax and CV separately; do not compare this with Table 6 median Tmax as though same summary.

Holt Study 2 (COV05100070)

Application
40 drops (approximately 1.2 mL) to EACH knee four times daily (Holt et al., 2015, Methods; Table 4, Study 2)
Population
Healthy adults aged 18–55; randomized open crossover; {'enrolled': 30, 'completers': 22}
Renal eligibility
Uncontrolled renal disease and abnormal renal laboratory tests excluded; numeric cutoff not given
Conditions
skin: clean, dry knees; major knee surgery excluded; heat: not reported; occlusion: not reported
Timing
7.5 days; steady-state results on day 8
AUC
353.78 ng·h/mL; mean; SD 174.93 (Holt et al., 2015, Table 4, Study 2)
Cmax
18.07 ng/mL; mean; SD 8.99 (Holt et al., 2015, Table 4, Study 2)
Tmax
Not reported in retrieved source (Holt et al., 2015, Table 4, Study 2)
Limitations
Oral comparator release-type conflict needs resolution before a selectable oral formulation is implemented. Results are Study 2 completers, not pooled with Studies 1 or 3. Drop volume/mass is study-specific; do not assume a universal drop-to-mL conversion.

Holt Study 2 (COV05100070)

Application
2 mL to EACH knee twice daily (Holt et al., 2015, Methods; Table 4, Study 2)
Population
Healthy adults aged 18–55; randomized open crossover; {'enrolled': 30, 'completers': 22}
Renal eligibility
Uncontrolled renal disease and abnormal renal laboratory tests excluded; numeric cutoff not given
Conditions
skin: clean, dry knees; major knee surgery excluded; heat: not reported; occlusion: not reported
Timing
7.5 days; steady-state results on day 8
AUC
315.47 ng·h/mL; mean; SD 158.7 (Holt et al., 2015, Table 4, Study 2)
Cmax
19.53 ng/mL; mean; SD 10.31 (Holt et al., 2015, Table 4, Study 2)
Tmax
Not reported in retrieved source (Holt et al., 2015, Table 4, Study 2)
Limitations
Oral comparator release-type conflict needs resolution before a selectable oral formulation is implemented. Results are Study 2 completers, not pooled with Studies 1 or 3. Drop volume/mass is study-specific; do not assume a universal drop-to-mL conversion.

Kienzler crossover, regimen A

Application
4 g to one knee four times daily (Kienzler et al., 2010, Abstract; historical US label §12.3, Table 2)
Population
Healthy adults; randomized 3-way crossover; {'enrolled': 40, 'completed_all_regimens': 39}
Renal eligibility
Healthy volunteers; detailed renal exclusion threshold not in abstract
Conditions
skin: not detailed in extracted study abstract; heat: separate US label experiment tested heat BEFORE application; not a correction factor; occlusion: not detailed in extracted study abstract
Timing
7 days; repeated-dose assessment
AUC
233 ng·h/mL; mean; SD 128 (Rebel Distributors Corp, 2010, §12.3, Table 2)
Cmax
15 ng/mL; mean; SD 7.3 (Rebel Distributors Corp, 2010, §12.3, Table 2)
Tmax
14 h; median; range 0–24 (Rebel Distributors Corp, 2010, §12.3, Table 2)
Limitations
US sodium gel is not Canadian diethylamine Emulgel. Reported percentage is retained as published; it is not recomputed from rounded arithmetic means. No prediction for other sites or quantities.

Kienzler crossover, regimen B

Application
4 g per knee to two knees + 2 g per hand to two hands, four times daily (Kienzler et al., 2010, Abstract; historical US label §12.3, Table 2)
Population
Healthy adults; randomized 3-way crossover; {'enrolled': 40, 'completed_all_regimens': 39}
Renal eligibility
Healthy volunteers; detailed renal exclusion threshold not in abstract
Conditions
skin: not detailed in extracted study abstract; heat: separate US label experiment tested heat BEFORE application; not a correction factor; occlusion: not detailed in extracted study abstract
Timing
7 days; repeated-dose assessment
AUC
807 ng·h/mL; mean; SD 478 (Rebel Distributors Corp, 2010, §12.3, Table 2)
Cmax
53.8 ng/mL; mean; SD 32 (Rebel Distributors Corp, 2010, §12.3, Table 2)
Tmax
10 h; median; range 0–24 (Rebel Distributors Corp, 2010, §12.3, Table 2)
Limitations
US sodium gel is not Canadian diethylamine Emulgel. Reported percentage is retained as published; it is not recomputed from rounded arithmetic means. 48 g/day arm exceeds the historical labeled 32 g/day maximum and is a research regimen, not a dosing recommendation.

Canadian monograph single-dose study

Application
1.0 mL once to one knee (Knight Therapeutics Inc., 2025, §10.3, pp. 20–21)
Population
Volunteers; health and renal eligibility not detailed; {'n': 6}
Renal eligibility
Not reported
Conditions
skin: not reported; heat: not reported; occlusion: not reported
Timing
Single application; urine collected to 120 h
AUC
Not reported in retrieved source (Knight Therapeutics Inc., 2025, pp. 20–21)
Cmax
9.7 ng/mL; mean; SD 4.7 (Knight Therapeutics Inc., 2025, pp. 20–21)
Tmax
24–48 h; reported interval (Knight Therapeutics Inc., 2025, pp. 20–21)
Limitations
Tmax reported as 24–48 h. Urinary recovery is not the percentage of oral AUC or a direct individual absorption estimate.

Flouvat repeated percutaneous dosing

Application
15 g once daily (Flouvat et al., 1989, Abstract)
Population
10 healthy young participants; {'n': 10}
Renal eligibility
Healthy participants; renal cutoff not supplied
Conditions
skin: not specified; no local intolerance reported; heat: not reported; occlusion: not reported
Timing
10 days; reported Cmax after first administration
AUC
Not reported in retrieved source (Flouvat et al., 1989, Abstract)
Cmax
144 ng/mL; mean; SD 91 (Flouvat et al., 1989, Abstract)
Tmax
Not reported in retrieved source (Flouvat et al., 1989, Abstract)
Limitations
Cmax is first administration, not a steady-state Cmax. No accumulation was reported at end of study; quantitative day-10 AUC/Cmax not supplied. Exact vehicle and brand not identified; do not map to Tiloket or InflaFlex.

Dowling PLO crossover

Application
Single 200 mg ketoprofen topical dose in 20% PLO (Dowling et al., 2004, Abstract)
Population
8 healthy volunteers; open randomized crossover; {'enrolled': 8, 'completed': 8}
Renal eligibility
Healthy volunteers; numeric renal eligibility not reported
Conditions
skin: not reported; heat: not reported; occlusion: not reported
Timing
Single dose; blood sampled through 10 h; 1-week washout
AUC
Not reported in retrieved source (Dowling et al., 2004, Abstract)
Cmax
0.021 µg/mL; median (Dowling et al., 2004, Abstract)
Tmax
Not reported in retrieved source (Dowling et al., 2004, Abstract)
Limitations
Original µg/mL units retained. Relative-bioavailability formula and AUC integration interval require full text; do not treat 0.48% as raw topical/oral AUC. No extrapolation from this 20% PLO to the 7% multi-ingredient compound.

Hui historical radiolabeled Pennsaid lotion

Application
Single topical radiolabeled 1.5% lotion application; amount not reported in abstract (Hui et al., 1998, Abstract)
Population
Human volunteers; sample size unavailable in abstract; {'n': None}
Renal eligibility
Not reported
Conditions
skin: not reported; heat: not reported; occlusion: not reported
Timing
24 h skin contact; then washed
AUC
Not reported in retrieved source (Hui et al., 1998, Abstract)
Cmax
Not reported in retrieved source (Hui et al., 1998, Abstract)
Tmax
Not reported in retrieved source (Hui et al., 1998, Abstract)
Limitations
Historical Dimethaid formulation; continuity with current product not fully verified. Radiolabel/metabolite recovery cannot be substituted for parent-diclofenac plasma AUC. Peak plasma 14C at 30 h; not labeled parent diclofenac Tmax.

InflaFlex: useful information without an invented exposure number

The provider brochure lists ketoprofen 7%, diclofenac 3%, amitriptyline 3%, baclofen 10%, lidocaine 4%, menthol 0.15% and camphor 0.15% in a proprietary transdermal base. Salt identities, percentage basis and complete vehicle are not established in the brochure. (InflaFlex, n.d., p. 2)

No exact-formulation human systemic PK study was located in this focused search. Show composition, provenance and “Systemic exposure not established.” Confirm the dispensed formula and vehicle with the compounding pharmacy before any active-mass arithmetic. Commercial diclofenac gels and the ketoprofen PLO study cannot supply an absorption fraction for this compound.

Renal considerations: keep the source categories separate

ProductLabel classificationAdjustment / monitoring / limitation
Voltaren Emulgel 1.16%No explicit numeric renal contraindication or renal dose adjustment identified in this monograph. (Haleon Canada ULC, 2026a, §§2, 4, 5, 7, 10.3; pp. 4–5, 7, 9, 23)The monograph states that no response/adverse-reaction laboratory monitoring is required and that effects of renal impairment on PK are unlikely to be clinically significant. These are label statements, not results of a dedicated CKD or post-AKI trial. Large-area/prolonged use may increase systemic effects. Renal failure reports also appear in the overdosage section. (Haleon Canada ULC, 2026a, §§2, 4, 5, 7, 10.3; pp. 4–5, 7, 9, 23)
Voltaren Emulgel Extra Strength 2.32%No explicit numeric renal contraindication or renal dose adjustment identified in this monograph. (Haleon Canada ULC, 2026b, §§2, 4, 7, 10.3; pp. 4–5, 9, 21–22)The monograph states no response/adverse-reaction laboratory monitoring is required, and renal-impairment effects on PK are unlikely to be clinically significant. No dedicated CKD/post-AKI PK study is supplied. Large-area/prolonged use may increase systemic effects. (Haleon Canada ULC, 2026b, §§2, 4, 7, 10.3; pp. 4–5, 9, 21–22)
Pennsaid 1.5%Explicit contraindication: severely impaired or deteriorating renal function (creatinine clearance <30 mL/min). Known hyperkalemia is also contraindicated. (Knight Therapeutics Inc., 2025, §2 p. 5; §7 pp. 8, 12–13)Lesser impairment requires monitoring. Consider lower doses with great caution in impaired renal function; no numerical renal adjustment. Consider SCr, CrCl, urea and electrolytes including potassium; periodic monitoring in those at hyperkalemia risk. Reduced perfusion/volume, heart failure, liver dysfunction, diuretics and older age increase concern. CrCl <60 mL/min (or 1 mL/sec) appears as a cardiovascular/renal risk factor, not as an additional absolute contraindication. (Knight Therapeutics Inc., 2025, §2 p. 5; §7 pp. 8, 12–13)
Tiloket 2.5% gelUse with caution in reduced renal function; no numeric threshold. (Tillomed Laboratories Ltd, 2023, §§4.4, 4.8)Isolated systemic renal reactions reported. No quantitative renal dose reduction or renal lab schedule supplied. Renal events are listed in postmarketing adverse effects. (Tillomed Laboratories Ltd, 2023, §§4.4, 4.8)
InflaFlex multi-ingredient compoundProduct-specific PK, renal contraindications, dose adjustments and monitoring not established. (InflaFlex, n.d., Composition only; focused search found no exact-formulation human PK)An information page is useful. Do not borrow commercial gel absorption, add NSAID percentages, or equate absence of data with low exposure. Other actives require separate clinical review. (InflaFlex, n.d., Composition only; focused search found no exact-formulation human PK)

Preserve the renal measure. Pennsaid’s cutoff is creatinine clearance <30 mL/min, not indexed eGFR. The wording also explicitly includes deteriorating renal function. Do not silently substitute eGFR or use an improving creatinine value as an automatic clearance to treat. (Knight Therapeutics Inc., 2025, §2, p. 5)

Clinical outcome evidence

EvidenceReported findingWhat it can support
Teo et al.: retrospective CKD cohort, n = 6298; eGFR <60 mL/min/1.73 m². (Teo et al., 2023, Abstract)30-day AKI: topical NSAIDs vs no NSAID adjusted OR 1.38 (95% CI 1.18–1.63); systemic vs none 1.77 (1.46–2.15). (Teo et al., 2023, Abstract)Association, not a causal or individual risk estimate. Mixed topical agents; not a precise diclofenac-product risk. Confounding and exposure ascertainment require full-text appraisal.
Lim et al.: retrospective cohort, n = 12,773; age ≥60. (Lim et al., 2021, Abstract)30-day AKI and/or hyperkalemia: topical vs none adjusted OR 1.48 (95% CI 1.31–1.67). In the CKD subgroup, topical vs short-course systemic OR 0.74 (0.60–0.90). (Lim et al., 2021, Abstract)Lower association than systemic treatment can coexist with higher association than no NSAID. Related setting/time period to Teo; do not pool as independent cohorts.
Roth & Fuller: pooled analysis of two 12-week randomized OA trials; n = 927. (Roth & Fuller, 2011, Abstract)Oral diclofenac produced greater creatinine increases and CrCl decreases than topical solution. (Roth & Fuller, 2011, Abstract)Supports a relative laboratory safety difference in selected trial populations; does not establish safety in CKD or immediately after AKI.
Krummel et al.: renal adverse-event case report. (Krummel et al., 2000, Case narrative)AKI following topical ketoprofen in a patient with CKD taking enalapril and furosemide; improvement after drug withdrawal. (Krummel et al., 2000, Case narrative)Shows a possible serious event. Cannot estimate incidence or isolate the effect of one concurrent medicine.

KDIGO supports medication review, individualized benefit–harm assessment and monitoring with potentially nephrotoxic medicines. Its discussion allows carefully supervised NSAID use in selected circumstances; it does not supply a topical “safe AUC,” a universal restart interval after AKI, or a product-specific risk calculator. (KDIGO CKD Work Group, 2024, Chapter 4, pp. S246–S247)

The 2026 renal-safety review was used for discovery and context. It highlights indirectness in high-risk populations and the need for diclofenac-specific CKD data. Its systematic-review designation and certainty assessments must not be attributed to this targeted review. (Coșovanu et al., 2026, Abstract)

A compact renal panel

Proposed fields: stable CKD / recent AKI / renal status unknown; trend improving / stable / worsening / unknown; volume depletion; ACE inhibitor or ARB; diuretic; another NSAID; and intended duration. Display exact product contraindications first, then relevant precautions and clinician-selected follow-up. These are review prompts, not a scored risk model.

Recent AKI remains a separate context even when renal function is improving. No reviewed source establishes a numerical exposure threshold or fixed recovery interval that makes treatment safe. Monitoring timing should remain a clinical decision rather than an invented default.

Keep nonrenal issues compact: NSAID hypersensitivity/ASA-sensitive asthma, GI bleeding or ulcer history, cardiovascular disease/heart failure, pregnancy, skin integrity and concurrent medicines. Product restrictions differ: Pennsaid contraindicates pregnancy in all trimesters, while the cited Canadian gel contraindication specifies the last trimester; earlier-pregnancy cautions remain in their respective monographs. Ketoprofen adds prominent photosensitivity precautions. (Knight Therapeutics Inc., 2025, §§2, 7) (Haleon Canada ULC, 2026a, §§2, 7) (Haleon Canada ULC, 2026b, §§2, 7) (Tillomed Laboratories Ltd, 2023, §§4.3–4.6)

Joints and studied application areas

Start with named joints and areas, without a painted body model. The latest design direction uses exact formulation and studied application choices. One knee, both knees, ankle and other study sites appear only where supported by the selected record. Custom side/rib and other areas remain visibly unestimated.

QuantiDerm’s existing calculation engine estimates FTUs and product supply. Its FTU-to-grams conversions, BSA adjustments and formulation factors will not supply NSAID doses or systemic exposure estimates. Doses remain tied to the exact product label or studied regimen. Optional model-based coverage estimates would need their own citations and assumptions.

For ribs or another custom site, record the site and actual amount; show “No matching studied regimen” and “Exposure not established.” Optional measured cm² may document treatment area. Do not convert that area into knee-equivalents, percentage absorbed or predicted AUC.

Feldmann and Maibach studied regional penetration of radiolabeled cortisol. That is relevant background for site differences, but it does not validate NSAID regional correction factors. No directly applicable human knee-to-rib multiplier for these exact formulations was identified. (Feldmann & Maibach, 1967)

A recent human diclofenac microdosing study found nonlinear systemic availability between microdose and therapeutic-dose conditions. This is additional reason to avoid assuming dose proportionality; it is not a correction formula for the commercial products. (Mairinger et al., 2025, Abstract)

The Canadian gel monographs describe effects of dose, area, contact time, hydration and occlusion. Such general statements do not validate an individualized site-by-area calculator. Keep heat, dressings and damaged skin as conditions that may invalidate the studied comparison—not as numeric multipliers. (Haleon Canada ULC, 2026a, §§7, 10.3) (Haleon Canada ULC, 2026b, §§7, 10.3)

What can be compared or calculated

ActionProposed rule
Display source resultsDefault. Keep original units, labels, summary statistic, variability and time window. AUC, Cmax, urinary recovery and bioavailability remain different measures.
Compare studied regimensPrefer the actual within-study comparator. Show absolute reported values together with reported ratios, if available. No cross-study ranking by a common percentage.
Calculate applied drug massOptional arithmetic only when the exact salt/concentration basis and quantity are verified. Mark “Calculated from label.” Applied mass is not absorbed mass. No assumed density or drop size.
Calculate a descriptive AUC ratioOptional evidence detail after clinical review, not a default or custom-dose calculator. Match analyte, time window and statistical summary. Retain the original published ratio separately.
Estimate an unstudied regimenNot established. No proportional scaling by amount, frequency, concentration, site or BSA.
Infer analgesic equivalence or renal safetyNot established. Do not calculate an oral-equivalent analgesic dose, combined NSAID burden, safe exposure cutoff or numerical kidney-risk score.
Worked distinction: a published percentage and a separate descriptive ratio

The Extra Strength source reports 4.5% dose-normalized relative bioavailability. Retain that wording as the reported result. (Haleon Canada ULC, 2026b, §10.3)

A separate calculation from the reported geometric means would be 67.9 ÷ 2830 × 100 ≈ 2.4%. This is the actual-regimen AUC ratio, without dose normalization. It does not replace 4.5%, estimate absorption from another site, or measure kidney risk. (Haleon Canada ULC, 2026b, Table 6)

Neither calculation is implemented in an app. The JSON labels this example as derived and keeps the reported measures intact.

A small, discoverable MVP

Use one primary screen: exact product → studied application → actual comparator → reported exposure → renal considerations. Keep the body picker optional. Do not ask for height, weight, BSA or laboratory values unless a reviewed feature actually uses them.

Design direction using Impeccable: familiar system typography, generous white space, strong contrast, restrained blue for actions, quiet dividers and clear labels. Critical restrictions stay visible. Avoid tooltip-only actions, icon-only navigation, oversized metric tiles and long onboarding text. The requested 80–90% discoverability is a design target to assess with pharmacists, not a measured result.

References: follow the useful patterns in your existing apps

PeritonitisCalc uses contextual evidence near results and a deduplicated expandable “References and evidence” footer. QuantiDerm uses linked source labels, an expandable methodology/reference note and a concise footer limitation. These local implementations were inspected; no changes were made to either app.

For this app, use a small author–year citation beside each dose and exposure value, linked to the matching APA entry and exact page/table. The footer should expose References · Evidence reviewed [date] · Version, with the full alphabetized APA list one click away. Use a shared source registry so a changed reference updates every citation.

Example citation treatment: 2 g morning and evening; maximum 4 g/day across all areas. (Haleon Canada ULC, 2026b, §4.2, p. 5)

The full reference is in the APA list below. The inline source carries the locator; the bibliographic entry carries the monograph revision date and stable link.

Staged development

  1. Clinical review now: agree on the Canadian core, confirm source extraction, resolve the Pennsaid oral-release discrepancy, and review the comparison rules and renal language.
  2. Small reference MVP: implement approved studied presets, original-unit displays, renal panel, InflaFlex information, inline citations and footer references. No inferred exposure for custom regimens.
  3. Optional body picker: add site selection if it improves speed in pharmacist testing. Add measured-area documentation only if useful; no exposure multiplier without new validation.
  4. Later oral module: a separate reference for product-specific oral regimens, precautions and indications may be worthwhile. Keep analgesic evidence, PK exposure and renal safety separate; do not introduce a universal NSAID equivalence table.

To control scope and download size, keep the initial evidence local, use simple joint/area choices, link to external source PDFs rather than embedding them, and avoid accounts, cloud services, an AI advice feature or a 3D body model.

References — APA 7

Alphabetized bibliography. Source-access dates and evidence locators are separate annotations, not part of the APA citation. Fixed dated monographs use their document revision/authorization date. Missing dates or DOIs are not invented.