Subcutaneous immunoglobulin (16 or 20%) therapy in obese patients with primary immunodeficiency: a retrospective analysis of administration by infusion pump or subcutaneous rapid push


Correspondence: R. S. Shapiro, 15700 37th Avenue No, Suite 110, Plymouth, MN 55446, USA.



A retrospective chart review was conducted at a single centre, capturing data on 173 primary immunodeficiency disease (PIDD) patients, including 40 obese patients, using subcutaneous administration of immunoglobulin (Ig) (SCIG) (16 or 20%) delivered by infusion pump or subcutaneous (s.c.) rapid push. Patients previously using Ig administered as intravenous (i.v.) infusions (IVIG) were converted to SCIG dosing on a 1:1 basis. In both obese and non-obese patients, mean serum Ig levels were higher during SCIG administration (steady state) compared with IVIG administration (trough values). Similar SCIG dose : serum IgG level relationships were observed between obese and non-obese patients, suggesting the consistent bioavailability of SCIG regardless of body mass index (BMI). The mean SCIG volume per dosing site and the mean number of dosing days per week were greater with s.c. rapid push compared with infusion pump in this cohort, but the mean number of sites per infusion session was lower with s.c. rapid push. Both methods were well tolerated. The use of 20 versus 16% SCIG in obese patients improved dosing efficiency, resulting in smaller weekly volumes (54·7 versus 74·5 ml/week) and dosing on fewer days per week (2·3 versus 3·4 days). These data do not suggest a need for SCIG dosing adjustments in obese individuals relative to non-obese patients. The administration of SCIG using either infusion pump or s.c. rapid push is a practical and well-tolerated alternative to IVIG in obese patients. Offering various administration techniques provides a greater opportunity for treatment satisfaction and patient empowerment, which may support high levels of patient compliance.


Immunoglobulin (Ig) replacement therapy in primary immunodeficiency disease (PIDD) has been administered traditionally as intravenous (i.v.) infusions (IVIG) [1]. Subcutaneous administration of Ig (SCIG) is gaining favour as an alternative route of administration with many potential benefits, including similar efficacy compared with IVIG [2-5], greater ease and convenience, more stable, steady-state serum IgG level profiles [6] and minimized risks of systemic adverse events from high peak values or ‘wearing-off’ effects from low IgG values, both of which can be associated with IVIG [2]. Switching from IVIG to SCIG has been associated with statistically significant improvements in quality of life and treatment satisfaction in children and adults with PIDD [7-9]. The conventional mode of SCIG administration has been via a programmable infusion pump. More recently, subcutaneous (s.c.) push using a syringe and a butterfly needle has emerged as an alternative technique that in many cases provides greater simplicity, more rapid drug delivery and greater convenience [10].

Clinical obesity presents unique challenges with regard to obtaining venous access for i.v. drug administration, and these patients are particularly appropriate candidates for s.c. administration of Ig. Not only does s.c. administration bypass the need for venous access, but patients with high body mass index (BMI) would probably be capable of administering high volumes of material subcutaneously [2]. Despite the high prevalence of obesity (>35% of adults and approximately 17% of children and adolescents in the United States) [11-13], there are few data available with regard to Ig dosing (i.v. or s.c.) in obese patients, as this population has generally been excluded from rigorous clinical trials. For intravenous dosing, it has been suggested to dose obese patients (with BMI ≥ 30 kg/m2 or those with body weight > 20% greater than ideal) according to adjusted body weight to account for increased distribution into extra body fluids [14], but the relevance of this recommendation to SCIG dosing is unclear. Based on area under the curve (AUC) equivalences, the findings of two SCIG pharmacokinetic studies suggested a possibly higher dose adjustment in patients with greater BMIs relative to other patients when switching from IVIG [1]. However, when the few most extreme (high and low) BMI patients were removed from the analysis, the difference was no longer statistically significant. To complicate the issue further, there remains some uncertainty about the appropriate i.v. : s.c. dosing conversion for all patients when transitioning from IVIG to SCIG. In the United States the recommended dose conversion ratio, when switching from IVIG to SCIG, is 1:1·37 for 16% SCIG and 1:1·53 for 20% SCIG, guidelines which are based upon AUC equivalencies [15, 16]. However, European regulators support a 1:1 dose conversion, requiring only that trough serum IgG levels on SCIG be at least comparable to those noted during prior therapy [17]. Our clinic's policy is to switch patients from IVIG to SCIG using a 1:1 dose conversion ratio, a practice that is supported by other published data [3, 5, 18, 19].

Our clinic has published retrospective chart review data previously on 104 PIDD patients voluntarily using SCIG via s.c. rapid push or infusion pump [10]. Ongoing data collection has expanded the database further to include a sufficiently large population of obese (BMI ≥ 30) patients to allow a subgroup analysis in this population. The primary objectives were to evaluate dosing and administration patterns and serum IgG levels associated with SCIG in obese individuals, and compare the safety and tolerability of s.c. rapid push and infusion pump techniques in these patients. A secondary objective was to examine the impact of the 20% SCIG preparation (Hizentra®; CSL Behring, King of Prussia, PA, USA) on dosing and administration patterns in obese patients with PIDD, many of whom had previously been using 16% SCIG (Vivaglobin®, CSL Behring).

Materials and methods


Individuals with a diagnosis of PIDD, referred to the Midwest Immunology Clinic (Plymouth, MN, USA) were included in the analysis. Eligible patients must have received at least one course of SCIG between January 2006 and November 2011. Patients were naive to Ig therapy or were being switched from IVIG to SCIG.

Eligible patients were offered a choice of SCIG administration via a portable, programmable infusion pump or s.c. rapid push, with both methods demonstrated to the patient before a decision was made. Patients were permitted to switch administration method at any time. Advanced practice nurses associated with the clinic provided thorough training on the selected method of administration to all patients, and caregivers when appropriate. Follow-up was conducted via monthly telephone contact as well as through clinic visits that generally occurred every 5–6 months. For volume considerations, our clinic uses higher concentration SCIG products (16%, 20%) versus 10% preferentially whenever possible; 10% SCIG is used only if a patient experiences a problem with the higher concentration product.

Data collection process

Demographic information and data relevant to each patient's SCIG treatment and previous IVIG regimen, if applicable, were recorded using a standard case report form. Clinical staff identified eligible patient charts, and data were abstracted and analysed by an independent clinical research company (Churchill Outcomes Research, Maplewood, NJ, USA). All patient information was anonymized in accordance with the Health Insurance Portability and Accountability Act Privacy Rule Section 164·514 and the Code of Federal Regulations, Title 45, Part 6, Protection of Human Subjects, and identifying patient information (e.g. names, identification numbers, medical record numbers, telephone numbers, addresses) was not retained or recorded. Data were reviewed and recorded descriptively. Each visit was captured and analysed according to the administration method and product in use at that time; thus, individual patients could migrate among treatment categories throughout the study. According to Chesapeake Research Review (Columbus, MD, USA), an independent institutional review board (IRB), the study met criteria for exemption from IRB oversight.

Infusion durations for s.c. rapid push dosing were extracted directly from patient charts, if noted. However, approximate administration times for the infusion pump needed to be estimated based upon recorded infusion site volumes and pump infusion rates. For infusion rates indicated as ranges (e.g. 15–19 ml/h), the average rate was used (17 ml/h in this example). In some instances, infusion rates were reported as a cut-off (e.g. < 5 ml/h or > 60 ml/h); for these calculations, the < and > signs were simply dropped, which may have led to slight over- or under-estimations of infusion times for some patients.


Study sample

The expanded data set included 173 patients, of whom 40 (23·1%) had a BMI ≥ 30 kg/m2 (‘obese’) (Table 1). Compared with lower BMI patients, the obese patients were typically older and predominantly female. Mean follow-up for all patients was 35·2 months (range 0·0–63·0 months).

Table 1. Patient demographics.
 BMI < 30 kg/m2BMI 30+ kg/m2
  1. At first visit. Individual body mass indices (BMIs) calculated as a mean across all visits. SCIG: s.c. immunoglobulin.
Age (years)  
Mean (range)20 (0–67)40 (3–67)
Mean (range)21·4 (12·0–29·9)38·8 (30·4–61·9)
Gender, n (%)
Male68 (51·1)12 (30·0)
Female65 (48·9)28 (70·0)
Initial chosen method of SCIG delivery, n (%)
Pump46 (34·6)14 (35·0)
Rapid push86 (64·7)26 (65·0)
Number of visits captured899241

IVIG versus SCIG

Among obese patients being switched from IVIG to SCIG, mean SCIG dosing was comparable, or slightly lower, on a g/kg/month basis relative to previous IVIG dosing (Fig. 1a). Mean weight-normalized monthly Ig doses (g/kg/month) were lower for the high-BMI versus lower-BMI patients for both IV and SC dosing, due in part probably to our clinic's general policy of initially ‘capping’ monthly Ig dosing at 80 g (ongoing dosing adjustments are made as necessary). In both obese and non-obese patients, mean serum IgG levels measured during SCIG therapy, which were more reflective of steady state conditions, were higher than trough levels measured during IVIG therapy in both BMI subgroups (Fig. 1b).

Figure 1.

Comparative immunoglobulin (Ig) dosing (a) and serum IgG findings (b) reflective of intravenous Ig (IVIG) and subcutaneous Ig (SCIG) use in patients with body mass index (BMI) < 30 and 30+. N: number of unique patients represented; V: number of visits with relevant data; s.d.: standard deviation

Infusion pump versus SC rapid push

The mean SCIG volume administered per dosing site was greater with s.c. rapid push versus infusion pump in both obese and non-obese patients (Table 2). However, the mean number of sites per infusion session was lower with the s.c. rapid push method compared with the infusion pump. The mean number of dosing days per week was about half a day greater with s.c. rapid push versus infusion pump (3·3 versus 2·7 days/week in obese patients, 2·7 versus 2·2 days/week in non-obese patients). Total mean weekly SCIG volume was greater among s.c. rapid push users compared with infusion pump users in both obese and non-obese cohorts.

Table 2. Dosing and administration patterns according to body mass index (BMI) status, immunoglobulin (Ig) route and subcutaneous (s.c.) Ig (SCIG) concentration.
 BMI < 30BMI 30+
  1. s.d.: standard deviation.
Infusion pump versus s.c. rapid push  
Age, mean, years (s.d.)  
Infusion pump16·9 (16·0)35·2 (17·9)
SC rapid push23·2 (18·8)41·9 (16·8)
Weight, mean, kg (s.d.)  
Infusion pump41·2 (27·6)77·0 (40·1)
SC rapid push47·7 (28·8)92·8 (40·8)
Mean SCIG dose, g/kg/month  
Infusion pump0·61 (0·20)0·53 (0·20)
SC rapid push0·58 (0·18)0·48 (0·16)
Weekly SCIG volume, ml per week, mean (s.d.)  
Infusion pump41·2 (24·2)62·4 (27·9)
SC rapid push46·2 (24·7)72·1 (31·7)
Dosing days per week, mean (s.d.)  
Infusion pump2·2 (0·9)2·7 (1·3)
SC rapid push2·7 (1·4)3·3 (1·5)
Sites per infusion session, mean (s.d.)  
Infusion pump1·7 (0·7)1·7 (0·9)
SC rapid push1·3 (0·6)1·5 (0·9)
SCIG volume per infusion site, ml, mean (s.d.)  
Infusion pump13·2 (6·3)16·8 (6·0)
SC rapid push15·1 (7·3)17·8 (8·3)
16% versus 20% SCIG  
Weekly SCIG volume, ml per week, mean (s.d.)  
16% SCIG45·4 (25·4)74·5 (32·0)
20% SCIG36·7 (19·0)54·7 (17·8)
Dosing days per week, mean (s.d.)  
16% SCIG2·66 (1·27)3·44 (1·46)
20% SCIG1·89 (0·86)2·30 (1·00)
Sites per infusion session, mean (s.d.)  
16% SCIG1·42 (0·65)1·65 (1·02)
20% SCIG1·53 (0·69)1·22 (0·42)
SCIG volume per infusion site, ml, mean (s.d.)  
16% SCIG14·1 (6·8)15·7 (5·7)
20% SCIG15·4 (8·1)23·1 (9·7)

Among obese patients, the mean monthly Ig dose during s.c. rapid push use was slightly lower than that noted during infusion pump use (0·48 versus 0·53 g/kg/month, respectively). However, mean serum IgG levels were higher during administration with s.c. rapid push versus infusion pump (997·0 versus 843·2 mg/dl). A majority of obese patients using the s.c. rapid push technique reported being able to administer their SCIG doses in 9 min or less, whether using 16% SCIG (60·7% of visits) or 20% SCIG (59·8% of visits) (Fig. 2). In comparison, among 42 obese patients with pump infusion rate data, the mean estimated duration of infusion was 55 min (median, 60 min; range 30–90 min).

Figure 2.

Time spent for subcutaneous (s.c.) rapid push infusion among obese patients [body mass index (BMI) ≥ 30]. Data are presented as the percentage of visits reporting administration times (in min) within each category. For comparison, all estimated infusion durations with pump administration were ≥ 30 min, regardless of s.c. immunoglobulin (SCIG) concentration (median, 60 min duration with both 16 and 20% SCIG).

Approximately 65% of all patients in each BMI subgroup selected s.c. rapid push as the initial method of SCIG delivery (Table 1). A greater percentage of obese patients who initially selected the push method either remained on the initial method or returned to the original method after switching (92·3%) compared with patients who initially selected the pump method (71·4%).

16 versus 20% SCIG

The use of 20% SCIG was associated with lower weekly product volumes compared to 16% SCIG, and this phenomenon was particularly robust among obese patients (Table 2). Relative to 16% SCIG, the use of 20% SCIG was also associated with a marked reduction in the mean number of weekly dosing days in both obese and non-obese patients. Among obese patients, the mean number of sites used per infusion session was lower during 20% SCIG use than with 16% SCIG. The mean volume per infusion site was similar for non-obese patients using 16 and 20% SCIG; however, among obese patients, the mean volume per infusion site was approximately one-third greater while using 20% SCIG compared with 16% SCIG (23·1 versus 15·7 ml). The maximum reported infusion site injection volume was 60 ml in the non-obese group and 50 ml in the obese group.

Of 1140 visits reporting data for this study, there were only two visits in which the use of 10% SCIG was noted (one patient in each BMI category, both using SCIG via push). In each instance, the patients were switched briefly to a 10% SCIG product following reports of local adverse reactions with the use of 16% SCIG. In each instance, the patient switched back to 16% SCIG at the following visit.

SCIG dose–response by BMI

Scatterplots correlating monthly weight-adjusted doses and serum IgG responses suggested nearly identical SCIG dose : serum IgG level relationships between high and low BMI patients (Fig. 3). Regression analysis confirmed no evidence of a relationship between serum IgG and BMI (P = 0·5379).

Figure 3.

Scatterplots illustrating serum immunoglobulin (IgG) responses according to monthly subcutaneous immunoglobulin (SCIG) doses for patients with body mass index (BMI) values < 30 and ≥ 30. Regression analysis confirmed no evidence of a relationship between serum IgG and BMI (P = 0·5379).

Safety and tolerability

Adverse event (AE) rates were low overall, the majority of events being local injection site reactions (Table 3). Despite the use of higher doses and greater weekly product volumes, overall AE rates were slightly lower among obese patients (15·8% of visits) compared to non-obese patients (17·6% of visits). The frequency of local reactions was also lower in the high BMI group compared with the lower BMI group (12·9% versus 15·0% of visits). When obese and non-obese patients were analysed together, overall AE rates were lower among patients using s.c. rapid push compared with the infusion pump (15·6 versus 20·7% of visits). The frequency of local reactions was also lower with s.c. rapid push versus the infusion pump (13·7 versus 17·6% of visits), despite a slightly greater mean SCIG volume per dosing site with s.c. rapid push in both obese and non-obese patients.

Table 3. Adverse events (AE).
AE, number of visits (%)BMI < 30 kg/m2BMI 30+ kg/m2
899 visits241 visits
  1. †More than one AE may have been reported at a visit. BMI, body mass index.
Any AE158 (17·6)38 (15·8)
Local reaction135 (15·0)31 (12·9)
Headache3 (0·3)6 (2·5)
Rash4 (0·4)2 (0·8)
Gastrointestinal2 (0·2)0 (0·0)
Other15 (1·7)2 (0·8)


This study captured the experience of a considerable number of obese patients with PIDD self-administering SCIG via s.c. rapid push or infusion pump. Data were collected on a total of 173 PIDD patients, 40 (23·1%) of whom were obese. The high prevalence of obesity in this patient sample is consistent with the 24·8% overall prevalence of obesity in Minnesota, where this study was conducted [11]. To our knowledge, this is the first such analysis of SCIG administration patterns in an obese PIDD population.

In the United States, general dose-conversion guidelines for switching from IVIG to SCIG recommend a 1:1·37 (16% SCIG) or 1:1·53 (20% SCIG) dose conversion ratio, based upon studies using AUC calculations for comparison [1, 15, 16]. However, AUC is not well established as a useful marker for correlating Ig pharmacokinetics with clinical outcomes [20], and these conversion ratios remain somewhat controversial. The maintenance of adequate trough values may be more clinically appropriate, as there is some evidence that trough IgG levels may correlate more accurately with clinical efficacy [1, 20, 21]. In the present study, dose conversion from IVIG to SCIG was generally on a 1:1 ratio, and mean serum IgG levels during SCIG therapy, representative of steady state conditions, were greater than 1000 mg/dl. Our results are supported by prior publications describing favourable pharmacokinetic outcomes in patients who were transitioned from IVIG to SCIG using a 1:1 dose-conversion ratio [3, 5, 18, 19]. Compared with non-obese patients, obese patients were dosed on a lower g/kg/month basis, due in part probably to the clinic policy of capping monthly Ig dosing at 80 g.

Our data, while not gathered as part of a rigorous pharmacokinetic study, seem to confirm that BMI status did not impact the serum Ig response to subcutaneously administered Ig, and this conclusion was supported by regression analysis findings. The proportional relationship between mean monthly weight-adjusted SCIG dose and corresponding mean serum IgG levels appeared to be consistent between high and low BMI groups. Furthermore, consistent bioavailability across BMI categories was suggested by similar dose–Ig response curves for obese and non-obese patients using SCIG.

In the current analysis, s.c. rapid push use was characterized by more frequent dosing on a weekly basis, but fewer infusion sites were used per dosing session. This is not surprising, as infusion pumps can be equipped to infuse more than one site at a time. The mean SCIG volume per dosing site was also slightly greater with s.c. rapid push versus infusion pump. There is no explanation as yet for the apparently higher serum IgG levels observed with the s.c. rapid push versus the infusion pump method of administration. This phenomenon may possibly be related to the slightly more frequent dosing intervals (days/week) evidenced during periods of s.c. rapid push usage. Rapid s.c. push use was associated with a higher mean total SCIG volume per week compared with the infusion pump. Analysis of group demographics revealed that push users were, on average, older and of greater absolute body weight compared with pump users, even within BMI subgroup categories. SCIG doses are predetermined for patients without regard to administration technique at our clinic; thus, it appears that patients with higher absolute dosing requirements tended to choose rapid s.c. push administration preferentially.

One clear advantage of the s.c. rapid push technique is time of administration. In our sample, a majority of obese patients were able to administer their SCIG doses in 9 min or less. In comparison, the median duration of infusion pump administration was 1 h, with times ranging from 30 to 90 min.

It should be noted that patients in our clinic have a certain amount of freedom in designing their weekly administration schedule. The medical staff determines the appropriate monthly dose, which is divided by 4 to give the required weekly dose. Thereafter, it is mainly left up to the patients regarding how they want to divide up the volume over the course of a week. Even though official dosing recommendations for SCIG recommend once weekly administration [15, 16], it is interesting to note that our population was typically dosing on multiple days per week by their own choice. While conventional wisdom might suggest that minimizing dosing frequency would be a very high priority with patients, it appears that there must be other, equally important factors. It is possible that the appeal of flexible, simply administered, relatively smaller-volume total doses is equally important to many patients, even if performed on a more frequent basis. Another interesting phenomenon was the observation that some patients were administering single-site volumes much higher than the recommended maximum volume of 25 ml per site [12], with 60 ml being the highest reported (18-year-old patient: BMI 24; 196 lbs). These findings suggest that higher-than-recommended injection volumes of 20% SCIG are well tolerated.

As expected, the availability of the higher concentration 20% SCIG product translated into more efficient dosing patterns compared with 16% SCIG. On average, total weekly SCIG volumes were lower during 20% SCIG use compared with 16% SCIG administration, with the volume discrepancy being particularly dramatic in the obese subset. For all patients using 20% SCIG, dosing was performed on fewer days per week, and using higher volumes per infusion site, relative to 16% SCIG use. This latter phenomenon was also more strongly pronounced among higher-BMI patients, perhaps because they were better able to tolerate larger volume injections.

While this retrospective study was not able to measure directly patient satisfaction with any of the administration techniques, an analysis of patient continuance with or switching of methods over a mean follow-up period of approximately 3 years provides a soft indication of these outcomes. A high degree of satisfaction with the s.c. rapid push method over the infusion pump method is suggested by the fact that, when offered a choice, 65% of patients chose to initiate SCIG therapy with this technique and 92·3% of these patients continued using s.c. rapid push or returned after trying the infusion pump. While both methods were well tolerated, the s.c. rapid push technique was associated specifically with a lower overall prevalence of all AEs and local site reactions. A small number of patients switched methods multiple times; our experience suggests that some patients alternate administration technique voluntarily, depending on life circumstances and which is most convenient at a given time.

Interpretation of these data is limited by the retrospective, observational nature of the study. Regardless, the real-world characteristics of the data and the sheer volume of visits captured add to the clinical value of these findings. This study was not designed to capture or analyse infection rates, thus our findings are also limited in that regard. However, the data reflect real-world management of PIDD patients who were being managed with the goal of infection minimization. In our study, the mean steady state IgG level in obese patients using SCIG therapy was 1069·5 mg/dl. While formal correlations relating Ig efficacy to pharmacokinetic variables (AUC, trough/steady-state levels, etc) do not exist, a recent meta-analysis of IVIG data reported that the pneumonia incidence among PIDD patients is reduced progressively up to a trough level of at least 1000 mg/dl [21]. A similar analysis of SCIG studies found a significant inverse correlation between annual infection rates and serum IgG concentrations [22].

While these findings provide, for the first time, data-supported evidence to help guide SCIG usage in obese patients, there can be considerable individual variation in serum IgG response in any patient. Dosing always needs to be individualized, with adjustments made according to serum IgG levels, clinical response and patient wellbeing [1, 2, 4, 23]. As a general rule, we ‘cap’ monthly Ig dosing initially in obese patients at a total of 80 g, but this is subject to ongoing adjustment as necessary.

In conclusion, a 1:1 dose switch when converting from IVIG to SCIG appears to produce proportional serum IgG levels among obese patients with PIDD. Our data do not suggest a need for unique dosing adjustments for SCIG in obese individuals. The 20% SCIG formulation allows for more efficient dosing of SCIG, which may be particularly advantageous in obese patients with greater absolute dosing requirements. Infusion pump and s.c. rapid push administration are both viable techniques for SCIG administration in obese patients, with no substantial measurable differences in drug delivery or safety and avoidance of the need for venous access. Ultimately, the choice of technique relies heavily upon patient preference. Having various administration techniques to offer patients provides a greater opportunity for treatment satisfaction and feelings of self-empowerment.


The author wishes to acknowledge Sandra Westra PharmD and Frank Rodino MHS, PA of Churchill Outcomes Research, for their assistance with data collection and analysis and editorial support, as well as Lisa Zacek of Midwest Immunology Clinic for her contributions to the data collection process. This study was funded through an unrestricted research grant from CSL Behring. The author maintained full control over the content of this manuscript throughout development.


The author serves as clinical investigator and advisor for CSL Behring, Baxter, Viropharma, Kedrion and as speaker for CSL, Baxter and Viropharma.