Low plasma viscosity as an indicator of antibody deficiency

Delay in the diagnosis of antibody immunodeficiency (low immunoglobulins) is common, with mean interval from first symptomatic presentation to diagnosis of 3.5 years in the UK. ¹ Delayed diagnosis is linked to significant and avoidable risk of bronchiectasis. Serendipitous laboratory clues pointing to antibody-related immunodeficiency include IgA deficiency in coeliac disease testing, ² low IgE in allergy testing ³ and low calculated globulin (total protein minus the albumin concentration^4,5).

A prolonged acute phase response elevates plasma viscosity (PV), and this assay is used primarily as a non-specific marker of inflammation. The major contributors to raised PV are fibrinogen and immunoglobulins. It has been shown that 57% of the total variation in PV is attributable to fibrinogen alone. ⁶ Next most important is IgG and then IgM. These three proteins contribute 72% of the total variation. ⁶ In contrast to a raised PV, the unexpected finding of a low PV (<1.5 cPa) is rarely investigated.

We undertook a pilot study, reviewing the data on patients with low PV over three months (1 January–April 30 2014) to test the hypothesis that low PV associates with low immunoglobulins and may be diagnostically useful. A total of 27,105 samples had a quantified PV; 1.8% (492 samples) had PV <1.5 cPa (reference range 1.5–1.72 cPa), with 256 a PV <1.48 cPa (0.94%). We reviewed laboratory data on the 256 samples with a PV <1.48 cPa (from 240 patients). Only 20 (8.3%) had had immunoglobulins (IgG, IgA and IgM) quantified within three weeks of the PV (one child, 19 adults). The child’s immunoglobulins were normal for age. Associated data records and stated clinical reason for testing were reviewed for the 19 adults.

The adult patients (10 female, nine male) were aged 16.0–91.2 years (mean 64.2). One or more immunoglobulins were low with respect to the adult (age 15–60 years) reference ranges in 14/19 (74%) patients (see Table 1).

Table 1.

Immunoglobulin balance in adult patients with low plasma viscosity.

		Patient	IgG g/L	IgA g/L	IgM g/L
Low IgG(nine patients)	Low IgA Low IgM	1	3.64	<0.25	<0.17
		2	4.68	<0.25	<0.17
	Low IgA Normal IgM	3	5.22	0.60	1.04
	Normal IgA Low IgM	4	5.62	1.29	0.23
		5	4.71	1.80	<0.16
		6 a	5.15	1.91	0.48
	Normal IgA Normal IgM	7	5.40	0.85	0.91
		8	5.15	1.06	0.97
		9	3.32	1.47	1.35
Normal IgG(10 patients)	Low IgA Low IgM	10a	6.62	0.60	0.40
	Low IgA Normal IgM
	Normal IgA Low IgM	11	6.00	2.23	0.26
		12	6.41	2.73	0.26
		13	6.43	1.47	0.30
		14	7.84	3.38	0.38
	Normal IgA Normal IgM	15	10.30	1.25	0.66
		16	6.40	1.45	0.70
		17	7.81	1.62	0.70
		18	8.33	1.79	0.77
		19	8.89	2.22	0.67

a

Two patients with low IgM actually had a normal IgM concentration when adjusted for an elderly adult reference range. ⁷

In 5/14 (36%) patients (3, 9, 1, 6 and 10 (see Table 1) in order of delay time) the first recorded low PV result preceded the first test for immunoglobulins. Delays were 0.25, 1.74, 3.31, 4.58 and 5.15 years (mean delay 3.01 years) from first low PV result to measurement of immunoglobulins. Clinical correlates for these five were immunosuppressive therapy (methotrexate), lymphoproliferative disease (two patients; light chain myeloma, monoclonal gammopathy of undetermined significance), persistent sore tongue with some Candida and persistent skin irritation.

Of interest, to verify the principle, we reviewed the data on our cohort of antibody-deficient patients. PV data were available pre-immunoglobulin replacement therapy in only six patients. PVs were 1.45, 1.46, 1.48, 1.52, 1.56 and 1.58 cPa. Whilst this is a very small cohort, the data suggest low PV is a common finding at presentation in antibody-deficient patients.

We note that there are limitations to this approach. The reasons for performing PV, i.e. inflammatory conditions, will bias the cohort and limit the pick up rate. Additionally, those who have low immunoglobulins but raised fibrinogen may have a normal PV. However, we do not claim that this approach will uncover all hypogammaglobulinaemic patients. We do contend that by not investigating further, we may miss cases that could be uncovered earlier in their clinical course.

In conclusion, these data support the hypothesis that low PV is frequently associated with low immunoglobulins. Associations include lymphoproliferative disorders, immunosuppressive therapy and otherwise unexplained hypogammaglobulinaemia. If these data are confirmed in larger studies, we would suggest that assessment of serum immunoglobulins should be considered in all patients with unexplained low PV, with the potential for early intervention.