- Email: [email protected]
Influenza vaccine and FDG-PET
CORRESPONDENCE
Author’s reply Sir—Ricardo Bastos makes the important point that physical examination is but one component of the clinical method. Done and interpreted in the light of other clinical information (most notably, the patient’s history), physical examination raises as many questions as it answers. Hence the importance of “rationalising” the entire clinical method, not merely the handson part. I agree with Bastos that more research is needed. Binu John and Smitha Thomas suggest that my reported incidence of pivotal physical findings might have been exaggerated because I was aware of previous examiners’ findings and patients’ responses to initial management. I agree that such information might have contextualised my approach to some patients, perhaps intensifying my focus on specific aspects of their examination. This information might have made my examination smarter, but it did not alter the fact that the findings were there to be found. The tricky methodological issue they and Bastos raise concerns the interdependent nature of the various types of information gleaned from the clinical method. One cannot quantify the effect of physical examination per se without studying a control group in whom physical examination is not done. In my view, such a study in sick inpatients is unethical. I hope John and Thomas will reconsider their opinion that physical examination soothes the psyche more than it reveals the soma. John Ioannidis raises the question of false positives and their potential to instigate unnecessary, potentially harmful, interventions. These outcomes comprised five of the 26 pivotal findings I reported. I cannot exclude the possibility that some of my own findings were also falsely positive. I found no evidence of this, but that is Ioannidis’s point: how could I, without an unbiased impeccable referee to certify my own accuracy? I can only say that I am more than willing to repeat my study if Ioannidis would kindly supply the referee. In addition, Ioannidis’s point about predictive values is well taken. But is he aware that “low risk” medical patients are not easy to find in US hospitals today? If I understand his point, then all the more reason to promote skilled physical examination in such settings. Finally, I am struck by the contrast between the letters of F J C Millard and John Ioannidis. I admire the clinical expertise of physicians in many developing countries who accomplish so much with so little technological
2024
activation. It might be very sensible for PET centres to check with their patients for recent immunisations, and delay imaging if this has occurred.
support. Unfortunately, lacking that support, some of them cannot verify the clinical outcomes that validate their expertise. I hope they can soon, as their countries develop. But I do not hope that their “development” extends to a point where they question seriously “whether whole subspecialties of our profession are useful”. In medicine, scepticism—not nihilism—stimulates progress.
D McCool, J R Buscombe, *A J W Hilson Department of Nuclear Medicine, Royal Free Hospital, London NW3 2QG, UK (e-mail: [email protected]) 1
Brendan Reilly Cook County Hospital, Room 2129, 1901 West Harrison Street, A-1502, Chicago, IL 60612, USA (e-mail: [email protected])
Iyengar S, Chin B, Margolick JB, Sabundayo BP, Schwartz DH. Anatomical loci of HIVassociated immune activation and association with viraemia. Lancet 2003; 362: 945–50.
Authors’ reply
Sir—D McCool and colleagues raise important points about the potential of recent vaccinations to confound diagnosis of metastases in regional nodes that drain injection sites. We did not generate a complete time-course of node activation, but did show strong signal 3 and 5 days after immunisation. This finding is remarkably consistent with the kinetics of lymphocyte activation and proliferation in well studied mouse models, in which cells undergo roughly 12 rounds of division within nodes, then move to the circulation 10–16 days after immunisation.1,2 It is also consistent with the recovery of recently activated lymphocytes in human circulation 14 days after immunisation. This said, there must be some variability among individuals, and between primary and anamnestic immunisations with regard to onset and duration of node activation. We want to define these kinetics but have no clear data at this time. Waiting at least 4 weeks after vaccination before using PET to diagnose tumour-involved nodes should minimise false positives from vaccine immunity, but might delay treatment. Repeat scanning at 2 and 4 weeks would be more informative, but logistically difficult. The situation can be even more confounding than the typical cases of vaccinees we described. Our first
Influenza vaccine and FDG-PET Sir—The paper by Sujatha Iyengar and colleagues (Sept 20, p 945)1 is fascinating in what it shows about the degree of metabolic activity produced in lymph nodes by viral infection. Although the major part of their work is on HIV, the fact that they showed that influenza vaccine produces lymphatic activation is of major clinical significance. The primary clinical use of fluorodeoxyglucose positron-emission tomography (FDG-PET) is in oncology, particularly for staging. This technique is most likely to be required in older age groups—the same patients who are most likely to have had recent influenza vaccine (indeed the UK government is encouraging this). The situation could therefore easily arise where an FDG-PET study is done on, say, a patient with carcinoma of the breast or lymphoma, who had had recent immunisation. Uptake identified in axillary lymph nodes might then be falsely assumed to be due to tumour. Clearly this would have a major effect on staging and subsequent therapy. Iyengar and colleagues do not give any time-scale for the duration of this
Coronal
R
L
R
L
Transaxial
R
L
R
L
FDG uptake in cervical axillary lymph nodes (two section depths) 5 days after first influenza vaccination Vaccination in left arm, FDG infusion in right.
THE LANCET • Vol 362 • December 13, 2003 • www.thelancet.com
For personal use. Only reproduce with permission from The Lancet publishing Group.
CORRESPONDENCE
influenza vaccinee was a healthy 28-year-old woman who, unlike the other volunteers, had not received yearly influenza vaccinations, but had a clinical history of a severe influenza-like syndrome 3 years before scanning. Her 5-day post-vaccination scan (figure) revealed extensive bilateral node activation throughout the chest external to the mediastinum. She was completely free of signs or symptoms during the ensuing extensive negative work-up for occult disease, and has remained completely healthy over the subsequent 6 years. We interpret this massive activation of upper torso nodes, in part, as a physical reflection of recall responses to non-vaccine strains—the “original antigenic sin” phenomenon, which is well described for influenza. It is also consistent with the finding that the secondary immune response to influenza infection exceeds the primary response by 1000-fold in terms of responding cell numbers.3 It hints at greater potential protection by natural infection than by yearly vaccination in the setting of a major antigenic shift in influenza. Furthermore, somewhat similar patterns of FDG uptake have been seen in some cancer patients, and clearly attributed to metabolically active brown fat.4,5 Such uptake can be present in the absence of tumour (unpublished data), and can be differentiated from lymph node or muscle activity by use of combined PET and computed tomography (PET/CT). Sujatha Iyengar, Bennett Chin, Beulah Sabundayo, Joseph Margolick, *David H Schwartz *Bloomberg School of Public Health, Department of Molecular Microbiology and Immunology (SI, BS, JM, DHS), and School of Medicine, Department of Radiology, Division of Nuclear Medicine (BC), Johns Hopkins University, Baltimore, MD 21205, USA (e-mail: [email protected]) Age (years)/sex/ occupation Patient number 1 33/M/doctor 2 28/F/nurse 3 26/F/nurse 4 45/F/nurse 5 26/F/nurse 6 49/F/register 7 51/F/porter 8 49/F/porter 9 34/M/doctor 10 53/F/doctor 11 41/F/nurse 12 27/F/nurse 13 35/F/nurse 14 28/F/nurse 15 30/F/doctor 16 42/M/technician 17 57/F/porter
1
2
3
4
5
Ahmed R, Gray D. Immunological memory and protective immunity: understanding their relation. Science 1996; 272: 54–60. Murali-Krishna K, Altman JD, Suresh M, Sourdive D, Zajac A, Ahmed R. In vivo dynamics of anti-viral CD8 T cell responses to different epitopes: an evaluation of bystander activation in primary and secondary responses to viral infection. Adv Exp Med Biol 1998; 452: 123–42. Flynn KJ, Belz GT, Altman JD, Woodland DL, Doherty PC. Virus-specific CD8+ T cells in primary and secondary influenza pneumonia. Immunity 1998; 8: 683–91. Hany TF, Gharehpapagh E, Kamel EM, Buck A, Himms-Hagen J, vonSchulthess GK. Brown adipose tissue: a factor to consider in symmetrical tracer uptake in the neck and upper chest region. Eur J Nucl Med 2002; 29: 1393–98. Cohade C, Mourtzkikos KA, Wahl RL. “USA-Fat”: prevalence is related to ambient outdoor temperature—evaluation with 18F-FDG PET/CT. J Nucl Med 2003; 44: 1267–70.
Treatment of severe acute respiratory syndrome in health-care workers Sir––There is still no proven therapy for severe acute respiratory syndrome (SARS). The protocol reported by Loletta So and colleagues1 emphasised the combination use of ribavirin and high-dose corticosteroid. However, if given very early in the course of the disease, this approach could suppress the generation of host immunity to the novel coronavirus. We observed a biphasic pattern of illness in SARS and postulated that the first stage represents a viraemic phase and the second phase is an immune phase.2 Acute respiratory distress syndrome seems to be a complication in the second phase.2 If this hypothesis is true, antiviral agents will be most beneficial in the first phase, whereas
corticosteroids should be delayed until the onset of the second phase to maximise benefit and keep the negative effects of immune suppression to a minimum. We, therefore, developed a treatment protocol3 that emphasises early use of ribavirin but delayed introduction of corticosteroids until the second week, if possible. One oral 2000 mg loading dose of ribavirin was given to patients as soon as they received medical attention, followed by 600 mg ribavirin twice daily for patients with a bodyweight greater than 75 kg or 1000 mg daily for those with a bodyweight of 75 kg or less (400 mg in morning, 600 mg in evening) for 10 days. For patients who developed pneumonia, intravenous methylprednisolone (2 mg/kg daily for 5 days) was started on day 8 of fever or later. If rapid deterioration occurred before day 8, steroid treatment was started upon development of dyspnoea. If respiratory distress was not responsive to this dose, 500 mg methylprednisolone daily for 3 days was given. Upon improvement, the dose of steroid was tapered off over the next 2 weeks as recovery warranted. In Taipei between April 23 and May 31, 2003, 17 health-care workers at our hospital contracted SARS.4 PCR of serum or nasopharyngeal swab proved positive in seven of the 17 health-care workers, and convalescent serum antibodies were positive in 13 health-care workers. All 17 health-care workers received our treatment protocol (table). The median starting day of ribavirin was day 2 of fever (range 1–7 days). Only one health-care worker needed subsequent intubation and respiratory support. All 17 individuals recovered without major sequelae or subsequent relapse. With prompt identification
Protection used
SARS category Ribavirin started*
Steroid started†
Minipulse started‡
ICU stay
Maximum oxygen demand
Discharge
N95 mask N95 mask N95 mask N95 mask N95 mask Surgical mask Surgical mask Surgical mask N95 mask N95 mask N95 mask N95 mask N95 mask N95 mask N95 mask N95 mask N95 mask
Probable Suspected Probable Probable Probable Probable Probable Probable Suspected Suspected Probable Suspected Probable Suspected Probable Probable Probable
D7 – D9 D10 D8 D6 D13 D7 – – D10 – D9 – D8 D11 D13
– – – – D13 D15 – – – – – – – – – – –
No No No No No Yes No No No No No No No No No No No
Room air Room air Room air Nasal cannula 3 L/min Nasal cannula 2 L/min Intubated with FiO2 0·6 Nasal cannula 3 L/min Nasal cannula 3 L/min Room air Room air Room air Room air Room air Room air Room air Nasal cannula 3 L/min Room air
D24 D18 D28 D27 D28 D52 D33 D25 D6 D14 D20 D10 D20 D12 D26 D29 D23
D1 D2 D3 D3 D2 D2 D7 D4 D6 D2 D1 D4 D2 D3 D2 D2 D3
F=female. M=male. D=day since onset of fever. ICU=intensive care unit; FiO2=fractional concentration of oxygen in inspired gas.*Oral ribavirin 2000 mg loading then 600 mg twice daily. †Intravenous methylprednisolone 2 mg/kg daily. ‡Intravenous methylprednisolone 500 mg daily.
Characteristics and timing of treatment for 17 health-care workers with SARS
THE LANCET • Vol 362 • December 13, 2003 • www.thelancet.com
2025
For personal use. Only reproduce with permission from The Lancet publishing Group.
Author’s reply Sir—Ricardo Bastos makes the important point that physical examination is but one component of the clinical method. Done and interpreted in the light of other clinical information (most notably, the patient’s history), physical examination raises as many questions as it answers. Hence the importance of “rationalising” the entire clinical method, not merely the handson part. I agree with Bastos that more research is needed. Binu John and Smitha Thomas suggest that my reported incidence of pivotal physical findings might have been exaggerated because I was aware of previous examiners’ findings and patients’ responses to initial management. I agree that such information might have contextualised my approach to some patients, perhaps intensifying my focus on specific aspects of their examination. This information might have made my examination smarter, but it did not alter the fact that the findings were there to be found. The tricky methodological issue they and Bastos raise concerns the interdependent nature of the various types of information gleaned from the clinical method. One cannot quantify the effect of physical examination per se without studying a control group in whom physical examination is not done. In my view, such a study in sick inpatients is unethical. I hope John and Thomas will reconsider their opinion that physical examination soothes the psyche more than it reveals the soma. John Ioannidis raises the question of false positives and their potential to instigate unnecessary, potentially harmful, interventions. These outcomes comprised five of the 26 pivotal findings I reported. I cannot exclude the possibility that some of my own findings were also falsely positive. I found no evidence of this, but that is Ioannidis’s point: how could I, without an unbiased impeccable referee to certify my own accuracy? I can only say that I am more than willing to repeat my study if Ioannidis would kindly supply the referee. In addition, Ioannidis’s point about predictive values is well taken. But is he aware that “low risk” medical patients are not easy to find in US hospitals today? If I understand his point, then all the more reason to promote skilled physical examination in such settings. Finally, I am struck by the contrast between the letters of F J C Millard and John Ioannidis. I admire the clinical expertise of physicians in many developing countries who accomplish so much with so little technological
2024
activation. It might be very sensible for PET centres to check with their patients for recent immunisations, and delay imaging if this has occurred.
support. Unfortunately, lacking that support, some of them cannot verify the clinical outcomes that validate their expertise. I hope they can soon, as their countries develop. But I do not hope that their “development” extends to a point where they question seriously “whether whole subspecialties of our profession are useful”. In medicine, scepticism—not nihilism—stimulates progress.
D McCool, J R Buscombe, *A J W Hilson Department of Nuclear Medicine, Royal Free Hospital, London NW3 2QG, UK (e-mail: [email protected]) 1
Brendan Reilly Cook County Hospital, Room 2129, 1901 West Harrison Street, A-1502, Chicago, IL 60612, USA (e-mail: [email protected])
Iyengar S, Chin B, Margolick JB, Sabundayo BP, Schwartz DH. Anatomical loci of HIVassociated immune activation and association with viraemia. Lancet 2003; 362: 945–50.
Authors’ reply
Sir—D McCool and colleagues raise important points about the potential of recent vaccinations to confound diagnosis of metastases in regional nodes that drain injection sites. We did not generate a complete time-course of node activation, but did show strong signal 3 and 5 days after immunisation. This finding is remarkably consistent with the kinetics of lymphocyte activation and proliferation in well studied mouse models, in which cells undergo roughly 12 rounds of division within nodes, then move to the circulation 10–16 days after immunisation.1,2 It is also consistent with the recovery of recently activated lymphocytes in human circulation 14 days after immunisation. This said, there must be some variability among individuals, and between primary and anamnestic immunisations with regard to onset and duration of node activation. We want to define these kinetics but have no clear data at this time. Waiting at least 4 weeks after vaccination before using PET to diagnose tumour-involved nodes should minimise false positives from vaccine immunity, but might delay treatment. Repeat scanning at 2 and 4 weeks would be more informative, but logistically difficult. The situation can be even more confounding than the typical cases of vaccinees we described. Our first
Influenza vaccine and FDG-PET Sir—The paper by Sujatha Iyengar and colleagues (Sept 20, p 945)1 is fascinating in what it shows about the degree of metabolic activity produced in lymph nodes by viral infection. Although the major part of their work is on HIV, the fact that they showed that influenza vaccine produces lymphatic activation is of major clinical significance. The primary clinical use of fluorodeoxyglucose positron-emission tomography (FDG-PET) is in oncology, particularly for staging. This technique is most likely to be required in older age groups—the same patients who are most likely to have had recent influenza vaccine (indeed the UK government is encouraging this). The situation could therefore easily arise where an FDG-PET study is done on, say, a patient with carcinoma of the breast or lymphoma, who had had recent immunisation. Uptake identified in axillary lymph nodes might then be falsely assumed to be due to tumour. Clearly this would have a major effect on staging and subsequent therapy. Iyengar and colleagues do not give any time-scale for the duration of this
Coronal
R
L
R
L
Transaxial
R
L
R
L
FDG uptake in cervical axillary lymph nodes (two section depths) 5 days after first influenza vaccination Vaccination in left arm, FDG infusion in right.
THE LANCET • Vol 362 • December 13, 2003 • www.thelancet.com
For personal use. Only reproduce with permission from The Lancet publishing Group.
CORRESPONDENCE
influenza vaccinee was a healthy 28-year-old woman who, unlike the other volunteers, had not received yearly influenza vaccinations, but had a clinical history of a severe influenza-like syndrome 3 years before scanning. Her 5-day post-vaccination scan (figure) revealed extensive bilateral node activation throughout the chest external to the mediastinum. She was completely free of signs or symptoms during the ensuing extensive negative work-up for occult disease, and has remained completely healthy over the subsequent 6 years. We interpret this massive activation of upper torso nodes, in part, as a physical reflection of recall responses to non-vaccine strains—the “original antigenic sin” phenomenon, which is well described for influenza. It is also consistent with the finding that the secondary immune response to influenza infection exceeds the primary response by 1000-fold in terms of responding cell numbers.3 It hints at greater potential protection by natural infection than by yearly vaccination in the setting of a major antigenic shift in influenza. Furthermore, somewhat similar patterns of FDG uptake have been seen in some cancer patients, and clearly attributed to metabolically active brown fat.4,5 Such uptake can be present in the absence of tumour (unpublished data), and can be differentiated from lymph node or muscle activity by use of combined PET and computed tomography (PET/CT). Sujatha Iyengar, Bennett Chin, Beulah Sabundayo, Joseph Margolick, *David H Schwartz *Bloomberg School of Public Health, Department of Molecular Microbiology and Immunology (SI, BS, JM, DHS), and School of Medicine, Department of Radiology, Division of Nuclear Medicine (BC), Johns Hopkins University, Baltimore, MD 21205, USA (e-mail: [email protected]) Age (years)/sex/ occupation Patient number 1 33/M/doctor 2 28/F/nurse 3 26/F/nurse 4 45/F/nurse 5 26/F/nurse 6 49/F/register 7 51/F/porter 8 49/F/porter 9 34/M/doctor 10 53/F/doctor 11 41/F/nurse 12 27/F/nurse 13 35/F/nurse 14 28/F/nurse 15 30/F/doctor 16 42/M/technician 17 57/F/porter
1
2
3
4
5
Ahmed R, Gray D. Immunological memory and protective immunity: understanding their relation. Science 1996; 272: 54–60. Murali-Krishna K, Altman JD, Suresh M, Sourdive D, Zajac A, Ahmed R. In vivo dynamics of anti-viral CD8 T cell responses to different epitopes: an evaluation of bystander activation in primary and secondary responses to viral infection. Adv Exp Med Biol 1998; 452: 123–42. Flynn KJ, Belz GT, Altman JD, Woodland DL, Doherty PC. Virus-specific CD8+ T cells in primary and secondary influenza pneumonia. Immunity 1998; 8: 683–91. Hany TF, Gharehpapagh E, Kamel EM, Buck A, Himms-Hagen J, vonSchulthess GK. Brown adipose tissue: a factor to consider in symmetrical tracer uptake in the neck and upper chest region. Eur J Nucl Med 2002; 29: 1393–98. Cohade C, Mourtzkikos KA, Wahl RL. “USA-Fat”: prevalence is related to ambient outdoor temperature—evaluation with 18F-FDG PET/CT. J Nucl Med 2003; 44: 1267–70.
Treatment of severe acute respiratory syndrome in health-care workers Sir––There is still no proven therapy for severe acute respiratory syndrome (SARS). The protocol reported by Loletta So and colleagues1 emphasised the combination use of ribavirin and high-dose corticosteroid. However, if given very early in the course of the disease, this approach could suppress the generation of host immunity to the novel coronavirus. We observed a biphasic pattern of illness in SARS and postulated that the first stage represents a viraemic phase and the second phase is an immune phase.2 Acute respiratory distress syndrome seems to be a complication in the second phase.2 If this hypothesis is true, antiviral agents will be most beneficial in the first phase, whereas
corticosteroids should be delayed until the onset of the second phase to maximise benefit and keep the negative effects of immune suppression to a minimum. We, therefore, developed a treatment protocol3 that emphasises early use of ribavirin but delayed introduction of corticosteroids until the second week, if possible. One oral 2000 mg loading dose of ribavirin was given to patients as soon as they received medical attention, followed by 600 mg ribavirin twice daily for patients with a bodyweight greater than 75 kg or 1000 mg daily for those with a bodyweight of 75 kg or less (400 mg in morning, 600 mg in evening) for 10 days. For patients who developed pneumonia, intravenous methylprednisolone (2 mg/kg daily for 5 days) was started on day 8 of fever or later. If rapid deterioration occurred before day 8, steroid treatment was started upon development of dyspnoea. If respiratory distress was not responsive to this dose, 500 mg methylprednisolone daily for 3 days was given. Upon improvement, the dose of steroid was tapered off over the next 2 weeks as recovery warranted. In Taipei between April 23 and May 31, 2003, 17 health-care workers at our hospital contracted SARS.4 PCR of serum or nasopharyngeal swab proved positive in seven of the 17 health-care workers, and convalescent serum antibodies were positive in 13 health-care workers. All 17 health-care workers received our treatment protocol (table). The median starting day of ribavirin was day 2 of fever (range 1–7 days). Only one health-care worker needed subsequent intubation and respiratory support. All 17 individuals recovered without major sequelae or subsequent relapse. With prompt identification
Protection used
SARS category Ribavirin started*
Steroid started†
Minipulse started‡
ICU stay
Maximum oxygen demand
Discharge
N95 mask N95 mask N95 mask N95 mask N95 mask Surgical mask Surgical mask Surgical mask N95 mask N95 mask N95 mask N95 mask N95 mask N95 mask N95 mask N95 mask N95 mask
Probable Suspected Probable Probable Probable Probable Probable Probable Suspected Suspected Probable Suspected Probable Suspected Probable Probable Probable
D7 – D9 D10 D8 D6 D13 D7 – – D10 – D9 – D8 D11 D13
– – – – D13 D15 – – – – – – – – – – –
No No No No No Yes No No No No No No No No No No No
Room air Room air Room air Nasal cannula 3 L/min Nasal cannula 2 L/min Intubated with FiO2 0·6 Nasal cannula 3 L/min Nasal cannula 3 L/min Room air Room air Room air Room air Room air Room air Room air Nasal cannula 3 L/min Room air
D24 D18 D28 D27 D28 D52 D33 D25 D6 D14 D20 D10 D20 D12 D26 D29 D23
D1 D2 D3 D3 D2 D2 D7 D4 D6 D2 D1 D4 D2 D3 D2 D2 D3
F=female. M=male. D=day since onset of fever. ICU=intensive care unit; FiO2=fractional concentration of oxygen in inspired gas.*Oral ribavirin 2000 mg loading then 600 mg twice daily. †Intravenous methylprednisolone 2 mg/kg daily. ‡Intravenous methylprednisolone 500 mg daily.
Characteristics and timing of treatment for 17 health-care workers with SARS
THE LANCET • Vol 362 • December 13, 2003 • www.thelancet.com
2025
For personal use. Only reproduce with permission from The Lancet publishing Group.