Patients over a wide anion gap range demonstrated no significant 
chloride bias from coulometric titrations. Bromide at 0.7 mmol/l gave 
a 
1 mmol/l chloride response (4).  Selectivity of chloride cartridge 
(mmol/l of anion required to increase chloride results by 1 mmol/l) 
(4).

A list of substances and conditions known to affect the level of 
sodium, potassium or chloride in vivo is given by both Young et al (5) 
and Friedman et al (6).  No representation is made by Roche 
Diagnostics 
regarding the completeness of these lists or the accuracy of the 
information contained therein.

Hemolysis: No significant interference up to an H index of 1000.

Icterus: No significant interference up to an I index of 60
(approximate conjugated and unconjugated bilirubin concentration: 60
mg/dL).

Lipemia: No significant interference up an L index of 2000.

References:
(1) Glick MR, Ryder KW, Glick SJ. Interferographs: Users Guide to
    Interferences in Clinical Chemistry Instruments, 2nd ed.
    Indianapolis, IN: Science Enterprises Inc; 1991:35-44.

(2) Tietz NW. Fundamentals of Clinical Chemistry. Philadelphia, Pa:
    3rd ed. WB Saunders Co; 1987:617.

(3) Tietz NW. Clinical Chemistry. Philadelphia, Pa: 2nd ed.,WB Saunders
    Co; 1994:1356, 1364, 1365.

(4) Simmons M, Sugahara K, Watanabe M. Development of an Improved
    Chloride Electrode for the Boehringer Mannheim/Hitachi 700 Series
    Analyzers. Poster session presented at the 1988 National Meeting
    of the American Association for Clinical Chemistry, New Orleans,
    LA, July 27, 1988 (reprints available from Roche Diagnostics
    Corporation).

(5) Young DS, et al. Clin Chem. 1975;21, No 5.

(6) Friedman RB, et al. Clin Chem. 1980;26, No 4.